Annotation of XML/HTMLparser.c, revision 1.3
1.1 daniel 1: /*
2: * HTMLparser.c : an HTML 4.0 non-verifying parser
3: *
4: * See Copyright for the status of this software.
5: *
6: * Daniel.Veillard@w3.org
7: */
8:
9: #ifdef WIN32
10: #define HAVE_FCNTL_H
11: #include <io.h>
12: #else
13: #include <config.h>
14: #endif
15: #include <stdio.h>
16: #include <ctype.h>
17: #include <string.h> /* for memset() only */
18: #include <stdlib.h>
19: #include <sys/stat.h>
20: #ifdef HAVE_FCNTL_H
21: #include <fcntl.h>
22: #endif
23: #ifdef HAVE_UNISTD_H
24: #include <unistd.h>
25: #endif
26: #ifdef HAVE_ZLIB_H
27: #include <zlib.h>
28: #endif
29:
30: #include "tree.h"
31: #include "HTMLparser.h"
32: #include "entities.h"
33: #include "encoding.h"
34: #include "valid.h"
35: #include "parserInternals.h"
36:
37: /* #define DEBUG */
38:
39: /************************************************************************
40: * *
41: * Parser stacks related functions and macros *
42: * *
43: ************************************************************************/
44:
45: /*
46: * Generic function for accessing stacks in the Parser Context
47: */
48:
49: #define PUSH_AND_POP(type, name) \
50: int html##name##Push(htmlParserCtxtPtr ctxt, type value) { \
51: if (ctxt->name##Nr >= ctxt->name##Max) { \
52: ctxt->name##Max *= 2; \
53: ctxt->name##Tab = (void *) realloc(ctxt->name##Tab, \
54: ctxt->name##Max * sizeof(ctxt->name##Tab[0])); \
55: if (ctxt->name##Tab == NULL) { \
56: fprintf(stderr, "realloc failed !\n"); \
57: exit(1); \
58: } \
59: } \
60: ctxt->name##Tab[ctxt->name##Nr] = value; \
61: ctxt->name = value; \
62: return(ctxt->name##Nr++); \
63: } \
64: type html##name##Pop(htmlParserCtxtPtr ctxt) { \
65: type ret; \
66: if (ctxt->name##Nr <= 0) return(0); \
67: ctxt->name##Nr--; \
68: if (ctxt->name##Nr > 0) \
69: ctxt->name = ctxt->name##Tab[ctxt->name##Nr - 1]; \
70: else \
71: ctxt->name = NULL; \
72: ret = ctxt->name##Tab[ctxt->name##Nr]; \
73: ctxt->name##Tab[ctxt->name##Nr] = 0; \
74: return(ret); \
75: } \
76:
77: PUSH_AND_POP(xmlNodePtr, node)
78:
79: /*
80: * Macros for accessing the content. Those should be used only by the parser,
81: * and not exported.
82: *
83: * Dirty macros, i.e. one need to make assumption on the context to use them
84: *
85: * CUR_PTR return the current pointer to the CHAR to be parsed.
86: * CUR returns the current CHAR value, i.e. a 8 bit value if compiled
87: * in ISO-Latin or UTF-8, and the current 16 bit value if compiled
88: * in UNICODE mode. This should be used internally by the parser
89: * only to compare to ASCII values otherwise it would break when
90: * running with UTF-8 encoding.
91: * NXT(n) returns the n'th next CHAR. Same as CUR is should be used only
92: * to compare on ASCII based substring.
93: * UPP(n) returns the n'th next CHAR converted to uppercase. Same as CUR
94: * it should be used only to compare on ASCII based substring.
95: * SKIP(n) Skip n CHAR, and must also be used only to skip ASCII defined
96: * strings within the parser.
97: *
98: * Clean macros, not dependent of an ASCII context, expect UTF-8 encoding
99: *
100: * CURRENT Returns the current char value, with the full decoding of
101: * UTF-8 if we are using this mode. It returns an int.
102: * NEXT Skip to the next character, this does the proper decoding
103: * in UTF-8 mode. It also pop-up unfinished entities on the fly.
104: * It returns the pointer to the current CHAR.
105: * COPY(to) copy one char to *to, increment CUR_PTR and to accordingly
106: */
107:
108: #define CUR (*ctxt->input->cur)
109: #define UPPER (toupper(*ctxt->input->cur))
110: #define SKIP(val) ctxt->input->cur += (val)
111: #define NXT(val) ctxt->input->cur[(val)]
112: #define UPP(val) (toupper(ctxt->input->cur[(val)]))
113: #define CUR_PTR ctxt->input->cur
114:
115: #define SKIP_BLANKS \
116: while (IS_BLANK(*(ctxt->input->cur))) NEXT
117:
118: #ifndef USE_UTF_8
119: #define CURRENT (*ctxt->input->cur)
120: #define NEXT ((*ctxt->input->cur) ? \
121: (((*(ctxt->input->cur) == '\n') ? \
122: (ctxt->input->line++, ctxt->input->col = 1) : \
123: (ctxt->input->col++)), ctxt->input->cur++) : \
124: (ctxt->input->cur))
125: #else
126: #endif
127:
128:
129: /************************************************************************
130: * *
131: * The list of HTML elements and their properties *
132: * *
133: ************************************************************************/
134:
135: /*
136: * Start Tag: 1 means the start tag can be ommited
137: * End Tag: 1 means the end tag can be ommited
138: * 2 means it's forbidden (empty elements)
139: * Depr: this element is deprecated
140: * DTD: 1 means that this element is valid only in the Loose DTD
141: * 2 means that this element is valid only in the Frameset DTD
142: *
143: * Name,Start Tag,End Tag, Empty, Depr., DTD, Description
144: */
145: htmlElemDesc html40ElementTable[] = {
146: { "A", 0, 0, 0, 0, 0, "anchor " },
147: { "ABBR", 0, 0, 0, 0, 0, "abbreviated form" },
148: { "ACRONYM", 0, 0, 0, 0, 0, "" },
149: { "ADDRESS", 0, 0, 0, 0, 0, "information on author " },
150: { "APPLET", 0, 0, 0, 1, 1, "Java applet " },
151: { "AREA", 0, 2, 1, 0, 0, "client-side image map area " },
152: { "B", 0, 0, 0, 0, 0, "bold text style" },
153: { "BASE", 0, 2, 1, 0, 0, "document base URI " },
154: { "BASEFONT", 0, 2, 1, 1, 1, "base font size " },
155: { "BDO", 0, 0, 0, 0, 0, "I18N BiDi over-ride " },
156: { "BIG", 0, 0, 0, 0, 0, "large text style" },
157: { "BLOCKQUOTE", 0, 0, 0, 0, 0, "long quotation " },
158: { "BODY", 1, 1, 0, 0, 0, "document body " },
159: { "BR", 0, 2, 1, 0, 0, "forced line break " },
160: { "BUTTON", 0, 0, 0, 0, 0, "push button " },
161: { "CAPTION", 0, 0, 0, 0, 0, "table caption " },
162: { "CENTER", 0, 0, 0, 1, 1, "shorthand for DIV align=center " },
163: { "CITE", 0, 0, 0, 0, 0, "citation" },
164: { "CODE", 0, 0, 0, 0, 0, "computer code fragment" },
165: { "COL", 0, 2, 1, 0, 0, "table column " },
166: { "COLGROUP", 0, 1, 0, 0, 0, "table column group " },
167: { "DD", 0, 1, 0, 0, 0, "definition description " },
168: { "DEL", 0, 0, 0, 0, 0, "deleted text " },
169: { "DFN", 0, 0, 0, 0, 0, "instance definition" },
170: { "DIR", 0, 0, 0, 1, 1, "directory list" },
171: { "DIV", 0, 0, 0, 0, 0, "generic language/style container"},
172: { "DL", 0, 0, 0, 0, 0, "definition list " },
173: { "DT", 0, 1, 0, 0, 0, "definition term " },
174: { "EM", 0, 0, 0, 0, 0, "emphasis" },
175: { "FIELDSET", 0, 0, 0, 0, 0, "form control group " },
176: { "FONT", 0, 0, 0, 1, 1, "local change to font " },
177: { "FORM", 0, 0, 0, 0, 0, "interactive form " },
178: { "FRAME", 0, 2, 1, 0, 2, "subwindow " },
179: { "FRAMESET", 0, 0, 0, 0, 2, "window subdivision" },
180: { "H1", 0, 0, 0, 0, 0, "heading " },
181: { "H2", 0, 0, 0, 0, 0, "heading " },
182: { "H3", 0, 0, 0, 0, 0, "heading " },
183: { "H4", 0, 0, 0, 0, 0, "heading " },
184: { "H5", 0, 0, 0, 0, 0, "heading " },
185: { "H6", 0, 0, 0, 0, 0, "heading " },
186: { "HEAD", 1, 1, 0, 0, 0, "document head " },
187: { "HR", 0, 2, 1, 0, 0, "horizontal rule " },
188: { "HTML", 1, 1, 0, 0, 0, "document root element " },
189: { "I", 0, 0, 0, 0, 0, "italic text style" },
190: { "IFRAME", 0, 0, 0, 0, 1, "inline subwindow " },
191: { "IMG", 0, 2, 1, 0, 0, "Embedded image " },
192: { "INPUT", 0, 2, 1, 0, 0, "form control " },
193: { "INS", 0, 0, 0, 0, 0, "inserted text" },
194: { "ISINDEX", 0, 2, 1, 1, 1, "single line prompt " },
195: { "KBD", 0, 0, 0, 0, 0, "text to be entered by the user" },
196: { "LABEL", 0, 0, 0, 0, 0, "form field label text " },
197: { "LEGEND", 0, 0, 0, 0, 0, "fieldset legend " },
198: { "LI", 0, 1, 0, 0, 0, "list item " },
199: { "LINK", 0, 2, 1, 0, 0, "a media-independent link " },
200: { "MAP", 0, 0, 0, 0, 0, "client-side image map " },
201: { "MENU", 0, 0, 0, 1, 1, "menu list " },
202: { "META", 0, 2, 1, 0, 0, "generic metainformation " },
203: { "NOFRAMES", 0, 0, 0, 0, 2, "alternate content container for non frame-based rendering " },
204: { "NOSCRIPT", 0, 0, 0, 0, 0, "alternate content container for non script-based rendering " },
205: { "OBJECT", 0, 0, 0, 0, 0, "generic embedded object " },
206: { "OL", 0, 0, 0, 0, 0, "ordered list " },
207: { "OPTGROUP", 0, 0, 0, 0, 0, "option group " },
208: { "OPTION", 0, 1, 0, 0, 0, "selectable choice " },
209: { "P", 0, 1, 0, 0, 0, "paragraph " },
210: { "PARAM", 0, 2, 1, 0, 0, "named property value " },
211: { "PRE", 0, 0, 0, 0, 0, "preformatted text " },
212: { "Q", 0, 0, 0, 0, 0, "short inline quotation " },
213: { "S", 0, 0, 0, 1, 1, "strike-through text style" },
214: { "SAMP", 0, 0, 0, 0, 0, "sample program output, scripts, etc." },
215: { "SCRIPT", 0, 0, 0, 0, 0, "script statements " },
216: { "SELECT", 0, 0, 0, 0, 0, "option selector " },
217: { "SMALL", 0, 0, 0, 0, 0, "small text style" },
218: { "SPAN", 0, 0, 0, 0, 0, "generic language/style container " },
219: { "STRIKE", 0, 0, 0, 1, 1, "strike-through text" },
220: { "STRONG", 0, 0, 0, 0, 0, "strong emphasis" },
221: { "STYLE", 0, 0, 0, 0, 0, "style info " },
222: { "SUB", 0, 0, 0, 0, 0, "subscript" },
223: { "SUP", 0, 0, 0, 0, 0, "superscript " },
224: { "TABLE", 0, 0, 0, 0, 0, " " },
225: { "TBODY", 1, 1, 0, 0, 0, "table body " },
226: { "TD", 0, 1, 0, 0, 0, "table data cell" },
227: { "TEXTAREA", 0, 0, 0, 0, 0, "multi-line text field " },
228: { "TFOOT", 0, 1, 0, 0, 0, "table footer " },
229: { "TH", 0, 1, 0, 0, 0, "table header cell" },
230: { "THEAD", 0, 1, 0, 0, 0, "table header " },
231: { "TITLE", 0, 0, 0, 0, 0, "document title " },
232: { "TR", 0, 1, 0, 0, 0, "table row " },
233: { "TT", 0, 0, 0, 0, 0, "teletype or monospaced text style" },
234: { "U", 0, 0, 0, 1, 1, "underlined text style" },
235: { "UL", 0, 0, 0, 0, 0, "unordered list " },
236: { "VAR", 0, 0, 0, 0, 0, "instance of a variable or program argument" },
237: };
238:
239: /*
240: * start tags that imply the end of a current element
241: * any tag of each line implies the end of the current element if the type of
242: * that element is in the same line
243: */
244: CHAR *htmlEquEnd[] = {
245: "DT", "DD", "LI", "OPTION", NULL,
246: "H1", "H2", "H3", "H4", "H5", "H6", NULL,
247: "OL", "MENU", "DIR", "ADDRESS", "PRE", "LISTING", "XMP", NULL,
248: NULL
249: };
250: /*
251: * acording the HTML DTD, HR should be added to the 2nd line above, as it
252: * is not allowed within a H1, H2, H3, etc. But we should tolerate that case
253: * because many documents contain rules in headings...
254: */
255:
256: /*
257: * start tags that imply the end of current element
258: */
259: CHAR *htmlStartClose[] = {
260: "FORM", "FORM", "P", "HR", "H1", "H2", "H3", "H4", "H5", "H6",
261: "DL", "UL", "OL", "MENU", "DIR", "ADDRESS", "PRE",
262: "LISTING", "XMP", "HEAD", NULL,
263: "HEAD", "P", NULL,
264: "TITLE", "P", NULL,
265: "BODY", "HEAD", "STYLE", "LINK", "TITLE", "P", NULL,
266: "LI", "P", "H1", "H2", "H3", "H4", "H5", "H6", "DL", "ADDRESS",
267: "PRE", "LISTING", "XMP", "HEAD", NULL,
268: "HR", "P", "HEAD", NULL,
269: "H1", "P", "HEAD", NULL,
270: "H2", "P", "HEAD", NULL,
271: "H3", "P", "HEAD", NULL,
272: "H4", "P", "HEAD", NULL,
273: "H5", "P", "HEAD", NULL,
274: "H6", "P", "HEAD", NULL,
275: "DIR", "P", "HEAD", NULL,
276: "ADDRESS", "P", "HEAD", "UL", NULL,
277: "PRE", "P", "HEAD", "UL", NULL,
278: "LISTING", "P", "HEAD", NULL,
279: "XMP", "P", "HEAD", NULL,
280: "BLOCKQUOTE", "P", "HEAD", NULL,
281: "DL", "P", "DT", "MENU", "DIR", "ADDRESS", "PRE", "LISTING",
282: "XMP", "HEAD", NULL,
283: "DT", "P", "MENU", "DIR", "ADDRESS", "PRE", "LISTING", "XMP", "HEAD", NULL,
284: "DD", "P", "MENU", "DIR", "ADDRESS", "PRE", "LISTING", "XMP", "HEAD", NULL,
285: "UL", "P", "HEAD", "OL", "MENU", "DIR", "ADDRESS", "PRE",
286: "LISTING", "XMP", NULL,
287: "OL", "P", "HEAD", "UL", NULL,
288: "MENU", "P", "HEAD", "UL", NULL,
289: "P", "P", "HEAD", "H1", "H2", "H3", "H4", "H5", "H6", NULL,
290: "DIV", "P", "HEAD", NULL,
291: "NOSCRIPT", "P", "HEAD", NULL,
292: "CENTER", "FONT", "B", "I", "P", "HEAD", NULL,
293: "A", "A", NULL,
294: "CAPTION", "P", NULL,
295: "COLGROUP", "CAPTION", "COLGROUP", "COL", "P", NULL,
296: "COL", "CAPTION", "COL", "P", NULL,
297: "TABLE", "P", "HEAD", "H1", "H2", "H3", "H4", "H5", "H6", "PRE",
298: "LISTING", "XMP", "A", NULL,
299: "TH", "TH", "TD", NULL,
300: "TD", "TH", "TD", NULL,
301: "TR", "TH", "TD", "TR", "CAPTION", "COL", "COLGROUP", NULL,
302: "THEAD", "CAPTION", "COL", "COLGROUP", NULL,
303: "TFOOT", "TH", "TD", "TR", "CAPTION", "COL", "COLGROUP", "THEAD",
304: "TBODY", NULL,
305: "TBODY", "TH", "TD", "TR", "CAPTION", "COL", "COLGROUP", "THEAD",
306: "TFOOT", "TBODY", NULL,
307: "OPTGROUP", "OPTION", NULL,
308: "FIELDSET", "LEGEND", "P", "HEAD", "H1", "H2", "H3", "H4", "H5", "H6",
309: "PRE", "LISTING", "XMP", "A", NULL,
310: NULL
311: };
312:
313: static CHAR** htmlStartCloseIndex[100];
314: static int htmlStartCloseIndexinitialized = 0;
315:
316: /************************************************************************
317: * *
318: * functions to handle HTML specific data *
319: * *
320: ************************************************************************/
321:
322: /**
323: * htmlInitAutoClose:
324: *
325: * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
326: *
327: */
328: void
329: htmlInitAutoClose(void) {
330: int index, i = 0;
331:
332: if (htmlStartCloseIndexinitialized) return;
333:
334: for (index = 0;index < 100;index ++) htmlStartCloseIndex[index] = NULL;
335: index = 0;
336: while ((htmlStartClose[i] != NULL) && (index < 100 - 1)) {
337: htmlStartCloseIndex[index++] = &htmlStartClose[i];
338: while (htmlStartClose[i] != NULL) i++;
339: i++;
340: }
341: }
342:
343: /**
344: * htmlTagLookup:
345: * @tag: The tag name
346: *
347: * Lookup the HTML tag in the ElementTable
348: *
349: * Returns the related htmlElemDescPtr or NULL if not found.
350: */
351: htmlElemDescPtr
352: htmlTagLookup(const CHAR *tag) {
353: int i = 0;
354:
355: for (i = 0; i < (sizeof(html40ElementTable) /
356: sizeof(html40ElementTable[0]));i++) {
357: if (!xmlStrcmp(tag, html40ElementTable[i].name))
358: return(&html40ElementTable[i]);
359: }
360: return(NULL);
361: }
362:
363: /**
364: * htmlCheckAutoClose:
365: * @new: The new tag name
366: * @old: The old tag name
367: *
368: * Checks wether the new tag is one of the registered valid tags for closing old.
369: * Initialize the htmlStartCloseIndex for fast lookup of closing tags names.
370: *
371: * Returns 0 if no, 1 if yes.
372: */
373: int
374: htmlCheckAutoClose(const CHAR *new, const CHAR *old) {
375: int i, index;
376: CHAR **close;
377:
378: if (htmlStartCloseIndexinitialized == 0) htmlInitAutoClose();
379:
380: /* inefficient, but not a big deal */
381: for (index = 0; index < 100;index++) {
382: close = htmlStartCloseIndex[index];
383: if (close == NULL) return(0);
384: if (!xmlStrcmp(*close, new)) break;
385: }
386:
387: i = close - htmlStartClose;
388: i++;
389: while (htmlStartClose[i] != NULL) {
390: if (!xmlStrcmp(htmlStartClose[i], old)) {
391: return(1);
392: }
393: i++;
394: }
395: return(0);
396: }
397:
398: /**
399: * htmlAutoClose:
400: * @ctxt: an HTML parser context
401: * @new: The new tag name
402: *
403: * The HTmL DtD allows a tag to implicitely close other tags.
404: * The list is kept in htmlStartClose array. This function is
405: * called when a new tag has been detected and generates the
406: * appropriates closes if possible/needed.
407: */
408: void
409: htmlAutoClose(htmlParserCtxtPtr ctxt, const CHAR *new) {
410:
411: while ((ctxt->node != NULL) &&
412: (htmlCheckAutoClose(new, ctxt->node->name))) {
413: #ifdef DEBUG
414: printf("htmlAutoClose: %s closes %s\n", new, ctxt->node->name);
415: #endif
416: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
417: ctxt->sax->endElement(ctxt->userData, ctxt->node->name);
418: }
419: }
420:
421: /**
422: * htmlAutoCloseOnClose:
423: * @ctxt: an HTML parser context
424: * @new: The new tag name
425: *
426: * The HTmL DtD allows an ending tag to implicitely close other tags.
427: */
428: void
429: htmlAutoCloseOnClose(htmlParserCtxtPtr ctxt, const CHAR *new) {
430: htmlElemDescPtr info;
431:
432: while ((ctxt->node != NULL) &&
433: (xmlStrcmp(new, ctxt->node->name))) {
434: info = htmlTagLookup(ctxt->node->name);
435: if ((info == NULL) || (info->endTag == 1)) {
436: #ifdef DEBUG
437: printf("htmlAutoCloseOnClose: %s closes %s\n", new, ctxt->node->name);
438: #endif
439: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
440: ctxt->sax->endElement(ctxt->userData, ctxt->node->name);
441: } else
442: break;
443: }
444: }
445:
446: /************************************************************************
447: * *
448: * The list of HTML predefined entities *
449: * *
450: ************************************************************************/
451:
452:
453: htmlEntityDesc html40EntitiesTable[] = {
454: /*
455: * the 4 absolute ones,
456: */
457: { 34, "quot", "quotation mark = APL quote, U+0022 ISOnum" },
458: { 38, "amp", "ampersand, U+0026 ISOnum" },
459: { 60, "lt", "less-than sign, U+003C ISOnum" },
460: { 62, "gt", "greater-than sign, U+003E ISOnum" },
461:
462: /*
463: * A bunch still in the 128-255 range
464: * Replacing them depend really on the charset used.
465: */
466: { 160, "nbsp", "no-break space = non-breaking space, U+00A0 ISOnum" },
467: { 161, "iexcl","inverted exclamation mark, U+00A1 ISOnum" },
468: { 162, "cent", "cent sign, U+00A2 ISOnum" },
469: { 163, "pound","pound sign, U+00A3 ISOnum" },
470: { 164, "curren","currency sign, U+00A4 ISOnum" },
471: { 165, "yen", "yen sign = yuan sign, U+00A5 ISOnum" },
472: { 166, "brvbar","broken bar = broken vertical bar, U+00A6 ISOnum" },
473: { 167, "sect", "section sign, U+00A7 ISOnum" },
474: { 168, "uml", "diaeresis = spacing diaeresis, U+00A8 ISOdia" },
475: { 169, "copy", "copyright sign, U+00A9 ISOnum" },
476: { 170, "ordf", "feminine ordinal indicator, U+00AA ISOnum" },
477: { 171, "laquo","left-pointing double angle quotation mark = left pointing guillemet, U+00AB ISOnum" },
478: { 172, "not", "not sign, U+00AC ISOnum" },
479: { 173, "shy", "soft hyphen = discretionary hyphen, U+00AD ISOnum" },
480: { 174, "reg", "registered sign = registered trade mark sign, U+00AE ISOnum" },
481: { 175, "macr", "macron = spacing macron = overline = APL overbar, U+00AF ISOdia" },
482: { 176, "deg", "degree sign, U+00B0 ISOnum" },
483: { 177, "plusmn","plus-minus sign = plus-or-minus sign, U+00B1 ISOnum" },
484: { 178, "sup2", "superscript two = superscript digit two = squared, U+00B2 ISOnum" },
485: { 179, "sup3", "superscript three = superscript digit three = cubed, U+00B3 ISOnum" },
486: { 180, "acute","acute accent = spacing acute, U+00B4 ISOdia" },
487: { 181, "micro","micro sign, U+00B5 ISOnum" },
488: { 182, "para", "pilcrow sign = paragraph sign, U+00B6 ISOnum" },
489: { 183, "middot","middle dot = Georgian comma
490: = Greek middle dot, U+00B7 ISOnum" },
491: { 184, "cedil","cedilla = spacing cedilla, U+00B8 ISOdia" },
492: { 185, "sup1", "superscript one = superscript digit one, U+00B9 ISOnum" },
493: { 186, "ordm", "masculine ordinal indicator, U+00BA ISOnum" },
494: { 187, "raquo","right-pointing double angle quotation mark
495: = right pointing guillemet, U+00BB ISOnum" },
496: { 188, "frac14","vulgar fraction one quarter = fraction one quarter, U+00BC ISOnum" },
497: { 189, "frac12","vulgar fraction one half = fraction one half, U+00BD ISOnum" },
498: { 190, "frac34","vulgar fraction three quarters = fraction three quarters, U+00BE ISOnum" },
499: { 191, "iquest","inverted question mark = turned question mark, U+00BF ISOnum" },
500: { 192, "Agrave","latin capital letter A with grave = latin capital letter A grave, U+00C0 ISOlat1" },
501: { 193, "Aacute","latin capital letter A with acute, U+00C1 ISOlat1" },
502: { 194, "Acirc","latin capital letter A with circumflex, U+00C2 ISOlat1" },
503: { 195, "Atilde","latin capital letter A with tilde, U+00C3 ISOlat1" },
504: { 196, "Auml", "latin capital letter A with diaeresis, U+00C4 ISOlat1" },
505: { 197, "Aring","latin capital letter A with ring above = latin capital letter A ring, U+00C5 ISOlat1" },
506: { 198, "AElig","latin capital letter AE = latin capital ligature AE, U+00C6 ISOlat1" },
507: { 199, "Ccedil","latin capital letter C with cedilla, U+00C7 ISOlat1" },
508: { 200, "Egrave","latin capital letter E with grave, U+00C8 ISOlat1" },
509: { 201, "Eacute","latin capital letter E with acute, U+00C9 ISOlat1" },
510: { 202, "Ecirc","latin capital letter E with circumflex, U+00CA ISOlat1" },
511: { 203, "Euml", "latin capital letter E with diaeresis, U+00CB ISOlat1" },
512: { 204, "Igrave","latin capital letter I with grave, U+00CC ISOlat1" },
513: { 205, "Iacute","latin capital letter I with acute, U+00CD ISOlat1" },
514: { 206, "Icirc","latin capital letter I with circumflex, U+00CE ISOlat1" },
515: { 207, "Iuml", "latin capital letter I with diaeresis, U+00CF ISOlat1" },
516: { 208, "ETH", "latin capital letter ETH, U+00D0 ISOlat1" },
517: { 209, "Ntilde","latin capital letter N with tilde, U+00D1 ISOlat1" },
518: { 210, "Ograve","latin capital letter O with grave, U+00D2 ISOlat1" },
519: { 211, "Oacute","latin capital letter O with acute, U+00D3 ISOlat1" },
520: { 212, "Ocirc","latin capital letter O with circumflex, U+00D4 ISOlat1" },
521: { 213, "Otilde","latin capital letter O with tilde, U+00D5 ISOlat1" },
522: { 214, "Ouml", "latin capital letter O with diaeresis, U+00D6 ISOlat1" },
523: { 215, "times","multiplication sign, U+00D7 ISOnum" },
524: { 216, "Oslash","latin capital letter O with stroke = latin capital letter O slash, U+00D8 ISOlat1" },
525: { 217, "Ugrave","latin capital letter U with grave, U+00D9 ISOlat1" },
526: { 218, "Uacute","latin capital letter U with acute, U+00DA ISOlat1" },
527: { 219, "Ucirc","latin capital letter U with circumflex, U+00DB ISOlat1" },
528: { 220, "Uuml", "latin capital letter U with diaeresis, U+00DC ISOlat1" },
529: { 221, "Yacute","latin capital letter Y with acute, U+00DD ISOlat1" },
530: { 222, "THORN","latin capital letter THORN, U+00DE ISOlat1" },
531: { 223, "szlig","latin small letter sharp s = ess-zed, U+00DF ISOlat1" },
532: { 224, "agrave","latin small letter a with grave = latin small letter a grave, U+00E0 ISOlat1" },
533: { 225, "aacute","latin small letter a with acute, U+00E1 ISOlat1" },
534: { 226, "acirc","latin small letter a with circumflex, U+00E2 ISOlat1" },
535: { 227, "atilde","latin small letter a with tilde, U+00E3 ISOlat1" },
536: { 228, "auml", "latin small letter a with diaeresis, U+00E4 ISOlat1" },
537: { 229, "aring","latin small letter a with ring above = latin small letter a ring, U+00E5 ISOlat1" },
538: { 230, "aelig","latin small letter ae = latin small ligature ae, U+00E6 ISOlat1" },
539: { 231, "ccedil","latin small letter c with cedilla, U+00E7 ISOlat1" },
540: { 232, "egrave","latin small letter e with grave, U+00E8 ISOlat1" },
541: { 233, "eacute","latin small letter e with acute, U+00E9 ISOlat1" },
542: { 234, "ecirc","latin small letter e with circumflex, U+00EA ISOlat1" },
543: { 235, "euml", "latin small letter e with diaeresis, U+00EB ISOlat1" },
544: { 236, "igrave","latin small letter i with grave, U+00EC ISOlat1" },
545: { 237, "iacute","latin small letter i with acute, U+00ED ISOlat1" },
546: { 238, "icirc","latin small letter i with circumflex, U+00EE ISOlat1" },
547: { 239, "iuml", "latin small letter i with diaeresis, U+00EF ISOlat1" },
548: { 240, "eth", "latin small letter eth, U+00F0 ISOlat1" },
549: { 241, "ntilde","latin small letter n with tilde, U+00F1 ISOlat1" },
550: { 242, "ograve","latin small letter o with grave, U+00F2 ISOlat1" },
551: { 243, "oacute","latin small letter o with acute, U+00F3 ISOlat1" },
552: { 244, "ocirc","latin small letter o with circumflex, U+00F4 ISOlat1" },
553: { 245, "otilde","latin small letter o with tilde, U+00F5 ISOlat1" },
554: { 246, "ouml", "latin small letter o with diaeresis, U+00F6 ISOlat1" },
555: { 247, "divide","division sign, U+00F7 ISOnum" },
556: { 248, "oslash","latin small letter o with stroke, = latin small letter o slash, U+00F8 ISOlat1" },
557: { 249, "ugrave","latin small letter u with grave, U+00F9 ISOlat1" },
558: { 250, "uacute","latin small letter u with acute, U+00FA ISOlat1" },
559: { 251, "ucirc","latin small letter u with circumflex, U+00FB ISOlat1" },
560: { 252, "uuml", "latin small letter u with diaeresis, U+00FC ISOlat1" },
561: { 253, "yacute","latin small letter y with acute, U+00FD ISOlat1" },
562: { 254, "thorn","latin small letter thorn with, U+00FE ISOlat1" },
563: { 255, "yuml", "latin small letter y with diaeresis, U+00FF ISOlat1" },
564:
565: /*
566: * Anything below should really be kept as entities references
567: */
568: { 402, "fnof", "latin small f with hook = function = florin, U+0192 ISOtech" },
569:
570: { 913, "Alpha","greek capital letter alpha, U+0391" },
571: { 914, "Beta", "greek capital letter beta, U+0392" },
572: { 915, "Gamma","greek capital letter gamma, U+0393 ISOgrk3" },
573: { 916, "Delta","greek capital letter delta, U+0394 ISOgrk3" },
574: { 917, "Epsilon","greek capital letter epsilon, U+0395" },
575: { 918, "Zeta", "greek capital letter zeta, U+0396" },
576: { 919, "Eta", "greek capital letter eta, U+0397" },
577: { 920, "Theta","greek capital letter theta, U+0398 ISOgrk3" },
578: { 921, "Iota", "greek capital letter iota, U+0399" },
579: { 922, "Kappa","greek capital letter kappa, U+039A" },
580: { 923, "Lambda""greek capital letter lambda, U+039B ISOgrk3" },
581: { 924, "Mu", "greek capital letter mu, U+039C" },
582: { 925, "Nu", "greek capital letter nu, U+039D" },
583: { 926, "Xi", "greek capital letter xi, U+039E ISOgrk3" },
584: { 927, "Omicron","greek capital letter omicron, U+039F" },
585: { 928, "Pi", "greek capital letter pi, U+03A0 ISOgrk3" },
586: { 929, "Rho", "greek capital letter rho, U+03A1" },
587: { 931, "Sigma","greek capital letter sigma, U+03A3 ISOgrk3" },
588: { 932, "Tau", "greek capital letter tau, U+03A4" },
589: { 933, "Upsilon","greek capital letter upsilon, U+03A5 ISOgrk3" },
590: { 934, "Phi", "greek capital letter phi, U+03A6 ISOgrk3" },
591: { 935, "Chi", "greek capital letter chi, U+03A7" },
592: { 936, "Psi", "greek capital letter psi, U+03A8 ISOgrk3" },
593: { 937, "Omega","greek capital letter omega, U+03A9 ISOgrk3" },
594:
595: { 945, "alpha","greek small letter alpha, U+03B1 ISOgrk3" },
596: { 946, "beta", "greek small letter beta, U+03B2 ISOgrk3" },
597: { 947, "gamma","greek small letter gamma, U+03B3 ISOgrk3" },
598: { 948, "delta","greek small letter delta, U+03B4 ISOgrk3" },
599: { 949, "epsilon","greek small letter epsilon, U+03B5 ISOgrk3" },
600: { 950, "zeta", "greek small letter zeta, U+03B6 ISOgrk3" },
601: { 951, "eta", "greek small letter eta, U+03B7 ISOgrk3" },
602: { 952, "theta","greek small letter theta, U+03B8 ISOgrk3" },
603: { 953, "iota", "greek small letter iota, U+03B9 ISOgrk3" },
604: { 954, "kappa","greek small letter kappa, U+03BA ISOgrk3" },
605: { 955, "lambda","greek small letter lambda, U+03BB ISOgrk3" },
606: { 956, "mu", "greek small letter mu, U+03BC ISOgrk3" },
607: { 957, "nu", "greek small letter nu, U+03BD ISOgrk3" },
608: { 958, "xi", "greek small letter xi, U+03BE ISOgrk3" },
609: { 959, "omicron","greek small letter omicron, U+03BF NEW" },
610: { 960, "pi", "greek small letter pi, U+03C0 ISOgrk3" },
611: { 961, "rho", "greek small letter rho, U+03C1 ISOgrk3" },
612: { 962, "sigmaf","greek small letter final sigma, U+03C2 ISOgrk3" },
613: { 963, "sigma","greek small letter sigma, U+03C3 ISOgrk3" },
614: { 964, "tau", "greek small letter tau, U+03C4 ISOgrk3" },
615: { 965, "upsilon","greek small letter upsilon, U+03C5 ISOgrk3" },
616: { 966, "phi", "greek small letter phi, U+03C6 ISOgrk3" },
617: { 967, "chi", "greek small letter chi, U+03C7 ISOgrk3" },
618: { 968, "psi", "greek small letter psi, U+03C8 ISOgrk3" },
619: { 969, "omega","greek small letter omega, U+03C9 ISOgrk3" },
620: { 977, "thetasym","greek small letter theta symbol, U+03D1 NEW" },
621: { 978, "upsih","greek upsilon with hook symbol, U+03D2 NEW" },
622: { 982, "piv", "greek pi symbol, U+03D6 ISOgrk3" },
623:
624: { 8226, "bull", "bullet = black small circle, U+2022 ISOpub" },
625: { 8230, "hellip","horizontal ellipsis = three dot leader, U+2026 ISOpub" },
626: { 8242, "prime","prime = minutes = feet, U+2032 ISOtech" },
627: { 8243, "Prime","double prime = seconds = inches, U+2033 ISOtech" },
628: { 8254, "oline","overline = spacing overscore, U+203E NEW" },
629: { 8260, "frasl","fraction slash, U+2044 NEW" },
630:
631: { 8472, "weierp","script capital P = power set
632: = Weierstrass p, U+2118 ISOamso" },
633: { 8465, "image","blackletter capital I = imaginary part, U+2111 ISOamso" },
634: { 8476, "real", "blackletter capital R = real part symbol, U+211C ISOamso" },
635: { 8482, "trade","trade mark sign, U+2122 ISOnum" },
636: { 8501, "alefsym","alef symbol = first transfinite cardinal, U+2135 NEW" },
637: { 8592, "larr", "leftwards arrow, U+2190 ISOnum" },
638: { 8593, "uarr", "upwards arrow, U+2191 ISOnum" },
639: { 8594, "rarr", "rightwards arrow, U+2192 ISOnum" },
640: { 8595, "darr", "downwards arrow, U+2193 ISOnum" },
641: { 8596, "harr", "left right arrow, U+2194 ISOamsa" },
642: { 8629, "crarr","downwards arrow with corner leftwards = carriage return, U+21B5 NEW" },
643: { 8656, "lArr", "leftwards double arrow, U+21D0 ISOtech" },
644: { 8657, "uArr", "upwards double arrow, U+21D1 ISOamsa" },
645: { 8658, "rArr", "rightwards double arrow, U+21D2 ISOtech" },
646: { 8659, "dArr", "downwards double arrow, U+21D3 ISOamsa" },
647: { 8660, "hArr", "left right double arrow, U+21D4 ISOamsa" },
648:
649:
650: { 8704, "forall","for all, U+2200 ISOtech" },
651: { 8706, "part", "partial differential, U+2202 ISOtech" },
652: { 8707, "exist","there exists, U+2203 ISOtech" },
653: { 8709, "empty","empty set = null set = diameter, U+2205 ISOamso" },
654: { 8711, "nabla","nabla = backward difference, U+2207 ISOtech" },
655: { 8712, "isin", "element of, U+2208 ISOtech" },
656: { 8713, "notin","not an element of, U+2209 ISOtech" },
657: { 8715, "ni", "contains as member, U+220B ISOtech" },
658: { 8719, "prod", "n-ary product = product sign, U+220F ISOamsb" },
659: { 8721, "sum", "n-ary sumation, U+2211 ISOamsb" },
660: { 8722, "minus","minus sign, U+2212 ISOtech" },
661: { 8727, "lowast","asterisk operator, U+2217 ISOtech" },
662: { 8730, "radic","square root = radical sign, U+221A ISOtech" },
663: { 8733, "prop", "proportional to, U+221D ISOtech" },
664: { 8734, "infin","infinity, U+221E ISOtech" },
665: { 8736, "ang", "angle, U+2220 ISOamso" },
666: { 8743, "and", "logical and = wedge, U+2227 ISOtech" },
667: { 8744, "or", "logical or = vee, U+2228 ISOtech" },
668: { 8745, "cap", "intersection = cap, U+2229 ISOtech" },
669: { 8746, "cup", "union = cup, U+222A ISOtech" },
670: { 8747, "int", "integral, U+222B ISOtech" },
671: { 8756, "there4","therefore, U+2234 ISOtech" },
672: { 8764, "sim", "tilde operator = varies with = similar to, U+223C ISOtech" },
673: { 8773, "cong", "approximately equal to, U+2245 ISOtech" },
674: { 8776, "asymp","almost equal to = asymptotic to, U+2248 ISOamsr" },
675: { 8800, "ne", "not equal to, U+2260 ISOtech" },
676: { 8801, "equiv","identical to, U+2261 ISOtech" },
677: { 8804, "le", "less-than or equal to, U+2264 ISOtech" },
678: { 8805, "ge", "greater-than or equal to, U+2265 ISOtech" },
679: { 8834, "sub", "subset of, U+2282 ISOtech" },
680: { 8835, "sup", "superset of, U+2283 ISOtech" },
681: { 8836, "nsub", "not a subset of, U+2284 ISOamsn" },
682: { 8838, "sube", "subset of or equal to, U+2286 ISOtech" },
683: { 8839, "supe", "superset of or equal to, U+2287 ISOtech" },
684: { 8853, "oplus","circled plus = direct sum, U+2295 ISOamsb" },
685: { 8855, "otimes","circled times = vector product, U+2297 ISOamsb" },
686: { 8869, "perp", "up tack = orthogonal to = perpendicular, U+22A5 ISOtech" },
687: { 8901, "sdot", "dot operator, U+22C5 ISOamsb" },
688: { 8968, "lceil","left ceiling = apl upstile, U+2308 ISOamsc" },
689: { 8969, "rceil","right ceiling, U+2309 ISOamsc" },
690: { 8970, "lfloor","left floor = apl downstile, U+230A ISOamsc" },
691: { 8971, "rfloor","right floor, U+230B ISOamsc" },
692: { 9001, "lang", "left-pointing angle bracket = bra, U+2329 ISOtech" },
693: { 9002, "rang", "right-pointing angle bracket = ket, U+232A ISOtech" },
694: { 9674, "loz", "lozenge, U+25CA ISOpub" },
695:
696: { 9824, "spades","black spade suit, U+2660 ISOpub" },
697: { 9827, "clubs","black club suit = shamrock, U+2663 ISOpub" },
698: { 9829, "hearts","black heart suit = valentine, U+2665 ISOpub" },
699: { 9830, "diams","black diamond suit, U+2666 ISOpub" },
700:
701: { 338, "OElig","latin capital ligature OE, U+0152 ISOlat2" },
702: { 339, "oelig","latin small ligature oe, U+0153 ISOlat2" },
703: { 352, "Scaron","latin capital letter S with caron, U+0160 ISOlat2" },
704: { 353, "scaron","latin small letter s with caron, U+0161 ISOlat2" },
705: { 376, "Yuml", "latin capital letter Y with diaeresis, U+0178 ISOlat2" },
706: { 710, "circ", "modifier letter circumflex accent, U+02C6 ISOpub" },
707: { 732, "tilde","small tilde, U+02DC ISOdia" },
708:
709: { 8194, "ensp", "en space, U+2002 ISOpub" },
710: { 8195, "emsp", "em space, U+2003 ISOpub" },
711: { 8201, "thinsp","thin space, U+2009 ISOpub" },
712: { 8204, "zwnj", "zero width non-joiner, U+200C NEW RFC 2070" },
713: { 8205, "zwj", "zero width joiner, U+200D NEW RFC 2070" },
714: { 8206, "lrm", "left-to-right mark, U+200E NEW RFC 2070" },
715: { 8207, "rlm", "right-to-left mark, U+200F NEW RFC 2070" },
716: { 8211, "ndash","en dash, U+2013 ISOpub" },
717: { 8212, "mdash","em dash, U+2014 ISOpub" },
718: { 8216, "lsquo","left single quotation mark, U+2018 ISOnum" },
719: { 8217, "rsquo","right single quotation mark, U+2019 ISOnum" },
720: { 8218, "sbquo","single low-9 quotation mark, U+201A NEW" },
721: { 8220, "ldquo","left double quotation mark, U+201C ISOnum" },
722: { 8221, "rdquo","right double quotation mark, U+201D ISOnum" },
723: { 8222, "bdquo","double low-9 quotation mark, U+201E NEW" },
724: { 8224, "dagger","dagger, U+2020 ISOpub" },
725: { 8225, "Dagger","double dagger, U+2021 ISOpub" },
726: { 8240, "permil","per mille sign, U+2030 ISOtech" },
727: { 8249, "lsaquo","single left-pointing angle quotation mark, U+2039 ISO proposed" },
728: { 8250, "rsaquo","single right-pointing angle quotation mark,
729: U+203A ISO proposed" },
730: { 8364, "euro", "euro sign, U+20AC NEW" }
731: };
732:
733: /************************************************************************
734: * *
735: * Commodity functions to handle entities *
736: * *
737: ************************************************************************/
738:
739: /*
740: * Macro used to grow the current buffer.
741: */
742: #define growBuffer(buffer) { \
743: buffer##_size *= 2; \
744: buffer = (CHAR *) realloc(buffer, buffer##_size * sizeof(CHAR)); \
745: if (buffer == NULL) { \
746: perror("realloc failed"); \
747: exit(1); \
748: } \
749: }
750:
751: /**
752: * htmlEntityLookup:
753: * @name: the entity name
754: *
755: * Lookup the given entity in EntitiesTable
756: *
757: * TODO: the linear scan is really ugly, an hash table is really needed.
758: *
759: * Returns the associated htmlEntityDescPtr if found, NULL otherwise.
760: */
761: htmlEntityDescPtr
762: htmlEntityLookup(const CHAR *name) {
763: int i;
764:
765: for (i = 0;i < (sizeof(html40EntitiesTable)/
766: sizeof(html40EntitiesTable[0]));i++) {
767: if (!xmlStrcmp(name, html40EntitiesTable[i].name)) {
768: #ifdef DEBUG
769: printf("Found entity %s\n", name);
770: #endif
771: return(&html40EntitiesTable[i]);
772: }
773: }
774: return(NULL);
775: }
776:
777:
778: /**
779: * htmlDecodeEntities:
780: * @ctxt: the parser context
781: * @len: the len to decode (in bytes !), -1 for no size limit
782: * @end: an end marker CHAR, 0 if none
783: * @end2: an end marker CHAR, 0 if none
784: * @end3: an end marker CHAR, 0 if none
785: *
786: * Subtitute the HTML entities by their value
787: *
788: * TODO: once the internal representation will be UTF-8, all entities
789: * will be substituable, in the meantime we only apply the substitution
790: * to the one with values in the 0-255 UNICODE range
791: *
792: * Returns A newly allocated string with the substitution done. The caller
793: * must deallocate it !
794: */
795: CHAR *
796: htmlDecodeEntities(htmlParserCtxtPtr ctxt, int len,
797: CHAR end, CHAR end2, CHAR end3) {
798: CHAR *buffer = NULL;
799: int buffer_size = 0;
800: CHAR *out = NULL;
801: CHAR *name = NULL;
802:
803: CHAR *cur = NULL;
804: htmlEntityDescPtr ent;
805: const CHAR *start = CUR_PTR;
806: unsigned int max = (unsigned int) len;
807:
808: /*
809: * allocate a translation buffer.
810: */
811: buffer_size = 1000;
812: buffer = (CHAR *) malloc(buffer_size * sizeof(CHAR));
813: if (buffer == NULL) {
814: perror("htmlDecodeEntities: malloc failed");
815: return(NULL);
816: }
817: out = buffer;
818:
819: /*
820: * Ok loop until we reach one of the ending char or a size limit.
821: */
822: while ((CUR_PTR - start < max) && (CUR != end) &&
823: (CUR != end2) && (CUR != end3)) {
824:
825: if (CUR == '&') {
826: if (NXT(1) == '#') {
827: int val = htmlParseCharRef(ctxt);
828: /* TODO: invalid for UTF-8 variable encoding !!! */
829: *out++ = val;
830: } else {
831: ent = htmlParseEntityRef(ctxt, &name);
832: if (name != NULL) {
833: if ((ent == NULL) || (ent->value <= 0) ||
834: (ent->value >= 255)) {
835: *out++ = '&';
836: cur = name;
837: while (*cur != 0) {
838: if (out - buffer > buffer_size - 100) {
839: int index = out - buffer;
840:
841: growBuffer(buffer);
842: out = &buffer[index];
843: }
844: *out++ = *cur++;
845: }
846: *out++ = ';';
847: } else {
848: /* TODO: invalid for UTF-8 variable encoding !!! */
849: *out++ = (CHAR)ent->value;
850: if (out - buffer > buffer_size - 100) {
851: int index = out - buffer;
852:
853: growBuffer(buffer);
854: out = &buffer[index];
855: }
856: }
857: free(name);
858: }
859: }
860: } else {
861: /* TODO: invalid for UTF-8 , use COPY(out); */
862: *out++ = CUR;
863: if (out - buffer > buffer_size - 100) {
864: int index = out - buffer;
865:
866: growBuffer(buffer);
867: out = &buffer[index];
868: }
869: NEXT;
870: }
871: }
872: *out++ = 0;
873: return(buffer);
874: }
875:
876:
877: /************************************************************************
878: * *
879: * Commodity functions to handle encodings *
880: * *
881: ************************************************************************/
882:
883: /**
884: * htmlSwitchEncoding:
885: * @ctxt: the parser context
886: * @len: the len of @cur
887: *
888: * change the input functions when discovering the character encoding
889: * of a given entity.
890: *
891: */
892: void
893: htmlSwitchEncoding(htmlParserCtxtPtr ctxt, xmlCharEncoding enc)
894: {
895: switch (enc) {
896: case XML_CHAR_ENCODING_ERROR:
897: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
898: ctxt->sax->error(ctxt->userData, "encoding unknown\n");
899: ctxt->wellFormed = 0;
900: break;
901: case XML_CHAR_ENCODING_NONE:
902: /* let's assume it's UTF-8 without the XML decl */
903: return;
904: case XML_CHAR_ENCODING_UTF8:
905: /* default encoding, no conversion should be needed */
906: return;
907: case XML_CHAR_ENCODING_UTF16LE:
908: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
909: ctxt->sax->error(ctxt->userData,
910: "char encoding UTF16 little endian not supported\n");
911: break;
912: case XML_CHAR_ENCODING_UTF16BE:
913: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
914: ctxt->sax->error(ctxt->userData,
915: "char encoding UTF16 big endian not supported\n");
916: break;
917: case XML_CHAR_ENCODING_UCS4LE:
918: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
919: ctxt->sax->error(ctxt->userData,
920: "char encoding USC4 little endian not supported\n");
921: break;
922: case XML_CHAR_ENCODING_UCS4BE:
923: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
924: ctxt->sax->error(ctxt->userData,
925: "char encoding USC4 big endian not supported\n");
926: break;
927: case XML_CHAR_ENCODING_EBCDIC:
928: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
929: ctxt->sax->error(ctxt->userData,
930: "char encoding EBCDIC not supported\n");
931: break;
932: case XML_CHAR_ENCODING_UCS4_2143:
933: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
934: ctxt->sax->error(ctxt->userData,
935: "char encoding UCS4 2143 not supported\n");
936: break;
937: case XML_CHAR_ENCODING_UCS4_3412:
938: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
939: ctxt->sax->error(ctxt->userData,
940: "char encoding UCS4 3412 not supported\n");
941: break;
942: case XML_CHAR_ENCODING_UCS2:
943: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
944: ctxt->sax->error(ctxt->userData,
945: "char encoding UCS2 not supported\n");
946: break;
947: case XML_CHAR_ENCODING_8859_1:
948: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
949: ctxt->sax->error(ctxt->userData,
950: "char encoding ISO_8859_1 ISO Latin 1 not supported\n");
951: break;
952: case XML_CHAR_ENCODING_8859_2:
953: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
954: ctxt->sax->error(ctxt->userData,
955: "char encoding ISO_8859_2 ISO Latin 2 not supported\n");
956: break;
957: case XML_CHAR_ENCODING_8859_3:
958: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
959: ctxt->sax->error(ctxt->userData,
960: "char encoding ISO_8859_3 not supported\n");
961: break;
962: case XML_CHAR_ENCODING_8859_4:
963: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
964: ctxt->sax->error(ctxt->userData,
965: "char encoding ISO_8859_4 not supported\n");
966: break;
967: case XML_CHAR_ENCODING_8859_5:
968: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
969: ctxt->sax->error(ctxt->userData,
970: "char encoding ISO_8859_5 not supported\n");
971: break;
972: case XML_CHAR_ENCODING_8859_6:
973: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
974: ctxt->sax->error(ctxt->userData,
975: "char encoding ISO_8859_6 not supported\n");
976: break;
977: case XML_CHAR_ENCODING_8859_7:
978: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
979: ctxt->sax->error(ctxt->userData,
980: "char encoding ISO_8859_7 not supported\n");
981: break;
982: case XML_CHAR_ENCODING_8859_8:
983: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
984: ctxt->sax->error(ctxt->userData,
985: "char encoding ISO_8859_8 not supported\n");
986: break;
987: case XML_CHAR_ENCODING_8859_9:
988: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
989: ctxt->sax->error(ctxt->userData,
990: "char encoding ISO_8859_9 not supported\n");
991: break;
992: case XML_CHAR_ENCODING_2022_JP:
993: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
994: ctxt->sax->error(ctxt->userData,
995: "char encoding ISO-2022-JPnot supported\n");
996: break;
997: case XML_CHAR_ENCODING_SHIFT_JIS:
998: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
999: ctxt->sax->error(ctxt->userData,
1000: "char encoding Shift_JISnot supported\n");
1001: break;
1002: case XML_CHAR_ENCODING_EUC_JP:
1003: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1004: ctxt->sax->error(ctxt->userData,
1005: "char encoding EUC-JPnot supported\n");
1006: break;
1007: }
1008: }
1009:
1010:
1011: /************************************************************************
1012: * *
1013: * Commodity functions, cleanup needed ? *
1014: * *
1015: ************************************************************************/
1016:
1017: /**
1018: * areBlanks:
1019: * @ctxt: an HTML parser context
1020: * @str: a CHAR *
1021: * @len: the size of @str
1022: *
1023: * Is this a sequence of blank chars that one can ignore ?
1024: *
1025: * TODO: to be corrected accodingly to DTD information if available
1026: *
1027: * Returns 1 if ignorable 0 otherwise.
1028: */
1029:
1030: static int areBlanks(htmlParserCtxtPtr ctxt, const CHAR *str, int len) {
1031: int i;
1032: xmlNodePtr lastChild;
1033:
1034: for (i = 0;i < len;i++)
1035: if (!(IS_BLANK(str[i]))) return(0);
1036:
1037: if (CUR != '<') return(0);
1038: if (ctxt->node == NULL) return(0);
1039: lastChild = xmlGetLastChild(ctxt->node);
1040: if (lastChild == NULL) {
1041: if (ctxt->node->content != NULL) return(0);
1042: } else if (xmlNodeIsText(lastChild))
1043: return(0);
1044: return(1);
1045: }
1046:
1047: /**
1048: * htmlHandleEntity:
1049: * @ctxt: an HTML parser context
1050: * @entity: an XML entity pointer.
1051: *
1052: * Default handling of an HTML entity, call the parser with the
1053: * substitution string
1054: */
1055:
1056: void
1057: htmlHandleEntity(htmlParserCtxtPtr ctxt, xmlEntityPtr entity) {
1058: int len;
1059:
1060: if (entity->content == NULL) {
1061: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1062: ctxt->sax->error(ctxt->userData, "htmlHandleEntity %s: content == NULL\n",
1063: entity->name);
1064: ctxt->wellFormed = 0;
1065: return;
1066: }
1067: len = xmlStrlen(entity->content);
1068:
1069: /*
1070: * Just handle the content as a set of chars.
1071: */
1072: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
1073: ctxt->sax->characters(ctxt->userData, entity->content, len);
1074:
1075: }
1076:
1077: /**
1078: * htmlNewDoc:
1079: * @URI: URI for the dtd, or NULL
1080: * @ExternalID: the external ID of the DTD, or NULL
1081: *
1082: * Returns a new document
1083: */
1084: htmlDocPtr
1085: htmlNewDoc(const CHAR *URI, const CHAR *ExternalID) {
1086: xmlDocPtr cur;
1087:
1088: /*
1089: * Allocate a new document and fill the fields.
1090: */
1091: cur = (xmlDocPtr) malloc(sizeof(xmlDoc));
1092: if (cur == NULL) {
1093: fprintf(stderr, "xmlNewDoc : malloc failed\n");
1094: return(NULL);
1095: }
1096:
1097: cur->type = XML_DOCUMENT_NODE;
1098: cur->version = NULL;
1099: cur->intSubset = NULL;
1100: xmlCreateIntSubset(cur, "HTML", ExternalID, URI);
1101: cur->name = NULL;
1102: cur->root = NULL;
1103: cur->extSubset = NULL;
1104: cur->oldNs = NULL;
1105: cur->encoding = NULL;
1106: cur->standalone = 1;
1107: cur->compression = 0;
1108: #ifndef XML_WITHOUT_CORBA
1109: cur->_private = NULL;
1110: cur->vepv = NULL;
1111: #endif
1112: return(cur);
1113: }
1114:
1115:
1116: /************************************************************************
1117: * *
1118: * The parser itself *
1119: * Relates to http://www.w3.org/TR/html40 *
1120: * *
1121: ************************************************************************/
1122:
1123: /************************************************************************
1124: * *
1125: * The parser itself *
1126: * *
1127: ************************************************************************/
1128:
1129: /**
1130: * htmlParseHTMLName:
1131: * @ctxt: an HTML parser context
1132: *
1133: * parse an HTML tag or attribute name, note that we convert it to uppercase
1134: * since HTML names are not case-sensitive.
1135: *
1136: * Returns the Tag Name parsed or NULL
1137: */
1138:
1139: CHAR *
1140: htmlParseHTMLName(htmlParserCtxtPtr ctxt) {
1141: CHAR *ret = NULL;
1142: int i = 0;
1143: CHAR loc[100];
1144:
1145: if (!IS_LETTER(CUR) && (CUR != '_') &&
1146: (CUR != ':')) return(NULL);
1147:
1148: while ((i < 100) && ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)))) {
1149: if ((CUR >= 0x61) && (CUR <= 0x7a)) loc[i] = CUR - 0x20;
1150: else loc[i] = CUR;
1151: i++;
1152:
1153: NEXT;
1154: }
1155:
1156: ret = xmlStrndup(loc, i);
1157:
1158: return(ret);
1159: }
1160:
1161: /**
1162: * htmlParseName:
1163: * @ctxt: an HTML parser context
1164: *
1165: * parse an HTML name, this routine is case sensistive.
1166: *
1167: * Returns the Name parsed or NULL
1168: */
1169:
1170: CHAR *
1171: htmlParseName(htmlParserCtxtPtr ctxt) {
1172: const CHAR *q;
1173: CHAR *ret = NULL;
1174:
1175: if (!IS_LETTER(CUR) && (CUR != '_') &&
1176: (CUR != ':')) return(NULL);
1177: q = NEXT;
1178:
1179: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1180: (CUR == '.') || (CUR == '-') ||
1181: (CUR == '_') || (CUR == ':') ||
1182: (IS_COMBINING(CUR)) ||
1183: (IS_EXTENDER(CUR)))
1184: NEXT;
1185:
1186: ret = xmlStrndup(q, CUR_PTR - q);
1187:
1188: return(ret);
1189: }
1190:
1191: /**
1192: * htmlParseHTMLAttribute:
1193: * @ctxt: an HTML parser context
1194: *
1195: * parse an HTML Nmtoken.
1196: *
1197: * Returns the Nmtoken parsed or NULL
1198: */
1199:
1200: CHAR *
1201: htmlParseHTMLAttribute(htmlParserCtxtPtr ctxt) {
1202: const CHAR *q;
1203: CHAR *ret = NULL;
1204:
1205: q = NEXT;
1206:
1207: while ((!IS_BLANK(CUR)) && (CUR != '<') &&
1208: (CUR != '&') && (CUR != '>') &&
1209: (CUR != '\'') && (CUR != '"'))
1210: NEXT;
1211:
1212: ret = xmlStrndup(q, CUR_PTR - q);
1213:
1214: return(ret);
1215: }
1216:
1217: /**
1218: * htmlParseNmtoken:
1219: * @ctxt: an HTML parser context
1220: *
1221: * parse an HTML Nmtoken.
1222: *
1223: * Returns the Nmtoken parsed or NULL
1224: */
1225:
1226: CHAR *
1227: htmlParseNmtoken(htmlParserCtxtPtr ctxt) {
1228: const CHAR *q;
1229: CHAR *ret = NULL;
1230:
1231: q = NEXT;
1232:
1233: while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) ||
1234: (CUR == '.') || (CUR == '-') ||
1235: (CUR == '_') || (CUR == ':') ||
1236: (IS_COMBINING(CUR)) ||
1237: (IS_EXTENDER(CUR)))
1238: NEXT;
1239:
1240: ret = xmlStrndup(q, CUR_PTR - q);
1241:
1242: return(ret);
1243: }
1244:
1245: /**
1246: * htmlParseEntityRef:
1247: * @ctxt: an HTML parser context
1248: * @str: location to store the entity name
1249: *
1250: * parse an HTML ENTITY references
1251: *
1252: * [68] EntityRef ::= '&' Name ';'
1253: *
1254: * Returns the associated htmlEntityDescPtr if found, or NULL otherwise,
1255: * if non-NULL *str will have to be freed by the caller.
1256: */
1257: htmlEntityDescPtr
1258: htmlParseEntityRef(htmlParserCtxtPtr ctxt, CHAR **str) {
1259: CHAR *name;
1260: htmlEntityDescPtr ent = NULL;
1261: *str = NULL;
1262:
1263: if (CUR == '&') {
1264: NEXT;
1265: name = htmlParseName(ctxt);
1266: if (name == NULL) {
1267: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1268: ctxt->sax->error(ctxt->userData, "htmlParseEntityRef: no name\n");
1269: ctxt->wellFormed = 0;
1270: } else {
1271: if (CUR == ';') {
1272: NEXT;
1273: *str = name;
1274:
1275: /*
1276: * Lookup the entity in the table.
1277: */
1278: ent = htmlEntityLookup(name);
1279: } else {
1280: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1281: ctxt->sax->error(ctxt->userData,
1282: "htmlParseEntityRef: expecting ';'\n");
1283: ctxt->wellFormed = 0;
1284: if (ctxt->sax->characters != NULL) {
1285: ctxt->sax->characters(ctxt->userData, "&", 1);
1286: ctxt->sax->characters(ctxt->userData, name, xmlStrlen(name));
1287: }
1288: free(name);
1289: }
1290: }
1291: }
1292: return(ent);
1293: }
1294:
1295: /**
1296: * htmlParseAttValue:
1297: * @ctxt: an HTML parser context
1298: *
1299: * parse a value for an attribute
1300: * Note: the parser won't do substitution of entities here, this
1301: * will be handled later in xmlStringGetNodeList, unless it was
1302: * asked for ctxt->replaceEntities != 0
1303: *
1304: * Returns the AttValue parsed or NULL.
1305: */
1306:
1307: CHAR *
1308: htmlParseAttValue(htmlParserCtxtPtr ctxt) {
1309: CHAR *ret = NULL;
1310:
1311: if (CUR == '"') {
1312: NEXT;
1313: ret = htmlDecodeEntities(ctxt, -1, '"', '<', 0);
1314: if (CUR == '<') {
1315: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1316: ctxt->sax->error(ctxt->userData,
1317: "Unescaped '<' not allowed in attributes values\n");
1318: ctxt->wellFormed = 0;
1319: }
1320: if (CUR != '"') {
1321: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1322: ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n");
1323: ctxt->wellFormed = 0;
1324: } else
1325: NEXT;
1326: } else if (CUR == '\'') {
1327: NEXT;
1328: ret = htmlDecodeEntities(ctxt, -1, '\'', '<', 0);
1329: if (CUR == '<') {
1330: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1331: ctxt->sax->error(ctxt->userData,
1332: "Unescaped '<' not allowed in attributes values\n");
1333: ctxt->wellFormed = 0;
1334: }
1335: if (CUR != '\'') {
1336: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1337: ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n");
1338: ctxt->wellFormed = 0;
1339: } else
1340: NEXT;
1341: } else {
1342: /*
1343: * That's an HTMLism, the attribute value may not be quoted
1344: */
1345: ret = htmlParseHTMLAttribute(ctxt);
1346: if (ret == NULL) {
1347: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1348: ctxt->sax->error(ctxt->userData, "AttValue: no value found\n");
1349: ctxt->wellFormed = 0;
1350: }
1351: }
1352:
1353: return(ret);
1354: }
1355:
1356: /**
1357: * htmlParseSystemLiteral:
1358: * @ctxt: an HTML parser context
1359: *
1360: * parse an HTML Literal
1361: *
1362: * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'")
1363: *
1364: * Returns the SystemLiteral parsed or NULL
1365: */
1366:
1367: CHAR *
1368: htmlParseSystemLiteral(htmlParserCtxtPtr ctxt) {
1369: const CHAR *q;
1370: CHAR *ret = NULL;
1371:
1372: if (CUR == '"') {
1373: NEXT;
1374: q = CUR_PTR;
1375: while ((IS_CHAR(CUR)) && (CUR != '"'))
1376: NEXT;
1377: if (!IS_CHAR(CUR)) {
1378: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1379: ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n");
1380: ctxt->wellFormed = 0;
1381: } else {
1382: ret = xmlStrndup(q, CUR_PTR - q);
1383: NEXT;
1384: }
1385: } else if (CUR == '\'') {
1386: NEXT;
1387: q = CUR_PTR;
1388: while ((IS_CHAR(CUR)) && (CUR != '\''))
1389: NEXT;
1390: if (!IS_CHAR(CUR)) {
1391: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1392: ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n");
1393: ctxt->wellFormed = 0;
1394: } else {
1395: ret = xmlStrndup(q, CUR_PTR - q);
1396: NEXT;
1397: }
1398: } else {
1399: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1400: ctxt->sax->error(ctxt->userData, "SystemLiteral \" or ' expected\n");
1401: ctxt->wellFormed = 0;
1402: }
1403:
1404: return(ret);
1405: }
1406:
1407: /**
1408: * htmlParsePubidLiteral:
1409: * @ctxt: an HTML parser context
1410: *
1411: * parse an HTML public literal
1412: *
1413: * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'"
1414: *
1415: * Returns the PubidLiteral parsed or NULL.
1416: */
1417:
1418: CHAR *
1419: htmlParsePubidLiteral(htmlParserCtxtPtr ctxt) {
1420: const CHAR *q;
1421: CHAR *ret = NULL;
1422: /*
1423: * Name ::= (Letter | '_') (NameChar)*
1424: */
1425: if (CUR == '"') {
1426: NEXT;
1427: q = CUR_PTR;
1428: while (IS_PUBIDCHAR(CUR)) NEXT;
1429: if (CUR != '"') {
1430: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1431: ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n");
1432: ctxt->wellFormed = 0;
1433: } else {
1434: ret = xmlStrndup(q, CUR_PTR - q);
1435: NEXT;
1436: }
1437: } else if (CUR == '\'') {
1438: NEXT;
1439: q = CUR_PTR;
1440: while ((IS_LETTER(CUR)) && (CUR != '\''))
1441: NEXT;
1442: if (!IS_LETTER(CUR)) {
1443: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1444: ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n");
1445: ctxt->wellFormed = 0;
1446: } else {
1447: ret = xmlStrndup(q, CUR_PTR - q);
1448: NEXT;
1449: }
1450: } else {
1451: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1452: ctxt->sax->error(ctxt->userData, "SystemLiteral \" or ' expected\n");
1453: ctxt->wellFormed = 0;
1454: }
1455:
1456: return(ret);
1457: }
1458:
1459: /**
1460: * htmlParseCharData:
1461: * @ctxt: an HTML parser context
1462: * @cdata: int indicating whether we are within a CDATA section
1463: *
1464: * parse a CharData section.
1465: * if we are within a CDATA section ']]>' marks an end of section.
1466: *
1467: * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*)
1468: */
1469:
1470: void
1471: htmlParseCharData(htmlParserCtxtPtr ctxt, int cdata) {
1472: const CHAR *q;
1473:
1474: q = CUR_PTR;
1475: while ((IS_CHAR(CUR)) && (CUR != '<') &&
1476: (CUR != '&')) {
1477: if ((CUR == ']') && (NXT(1) == ']') &&
1478: (NXT(2) == '>')) {
1479: if (cdata) break;
1480: else {
1481: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1482: ctxt->sax->error(ctxt->userData,
1483: "Sequence ']]>' not allowed in content\n");
1484: ctxt->wellFormed = 0;
1485: }
1486: }
1487: NEXT;
1488: }
1489: if (q == CUR_PTR) return;
1490:
1491: /*
1492: * Ok the segment [q CUR_PTR] is to be consumed as chars.
1493: */
1494: if (ctxt->sax != NULL) {
1495: if (areBlanks(ctxt, q, CUR_PTR - q)) {
1496: if (ctxt->sax->ignorableWhitespace != NULL)
1497: ctxt->sax->ignorableWhitespace(ctxt->userData, q, CUR_PTR - q);
1498: } else {
1499: if (ctxt->sax->characters != NULL)
1500: ctxt->sax->characters(ctxt->userData, q, CUR_PTR - q);
1501: }
1502: }
1503: }
1504:
1505: /**
1506: * htmlParseExternalID:
1507: * @ctxt: an HTML parser context
1508: * @publicID: a CHAR** receiving PubidLiteral
1509: * @strict: indicate whether we should restrict parsing to only
1510: * production [75], see NOTE below
1511: *
1512: * Parse an External ID or a Public ID
1513: *
1514: * NOTE: Productions [75] and [83] interract badly since [75] can generate
1515: * 'PUBLIC' S PubidLiteral S SystemLiteral
1516: *
1517: * [75] ExternalID ::= 'SYSTEM' S SystemLiteral
1518: * | 'PUBLIC' S PubidLiteral S SystemLiteral
1519: *
1520: * [83] PublicID ::= 'PUBLIC' S PubidLiteral
1521: *
1522: * Returns the function returns SystemLiteral and in the second
1523: * case publicID receives PubidLiteral, is strict is off
1524: * it is possible to return NULL and have publicID set.
1525: */
1526:
1527: CHAR *
1528: htmlParseExternalID(htmlParserCtxtPtr ctxt, CHAR **publicID, int strict) {
1529: CHAR *URI = NULL;
1530:
1531: if ((UPPER == 'S') && (UPP(1) == 'Y') &&
1532: (UPP(2) == 'S') && (UPP(3) == 'T') &&
1533: (UPP(4) == 'E') && (UPP(5) == 'M')) {
1534: SKIP(6);
1535: if (!IS_BLANK(CUR)) {
1536: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1537: ctxt->sax->error(ctxt->userData,
1538: "Space required after 'SYSTEM'\n");
1539: ctxt->wellFormed = 0;
1540: }
1541: SKIP_BLANKS;
1542: URI = htmlParseSystemLiteral(ctxt);
1543: if (URI == NULL) {
1544: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1545: ctxt->sax->error(ctxt->userData,
1546: "htmlParseExternalID: SYSTEM, no URI\n");
1547: ctxt->wellFormed = 0;
1548: }
1549: } else if ((UPPER == 'P') && (UPP(1) == 'U') &&
1550: (UPP(2) == 'B') && (UPP(3) == 'L') &&
1551: (UPP(4) == 'I') && (UPP(5) == 'C')) {
1552: SKIP(6);
1553: if (!IS_BLANK(CUR)) {
1554: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1555: ctxt->sax->error(ctxt->userData,
1556: "Space required after 'PUBLIC'\n");
1557: ctxt->wellFormed = 0;
1558: }
1559: SKIP_BLANKS;
1560: *publicID = htmlParsePubidLiteral(ctxt);
1561: if (*publicID == NULL) {
1562: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1563: ctxt->sax->error(ctxt->userData,
1564: "htmlParseExternalID: PUBLIC, no Public Identifier\n");
1565: ctxt->wellFormed = 0;
1566: }
1567: if (strict) {
1568: /*
1569: * We don't handle [83] so "S SystemLiteral" is required.
1570: */
1571: if (!IS_BLANK(CUR)) {
1572: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1573: ctxt->sax->error(ctxt->userData,
1574: "Space required after the Public Identifier\n");
1575: ctxt->wellFormed = 0;
1576: }
1577: } else {
1578: /*
1579: * We handle [83] so we return immediately, if
1580: * "S SystemLiteral" is not detected. From a purely parsing
1581: * point of view that's a nice mess.
1582: */
1583: const CHAR *ptr = CUR_PTR;
1584: if (!IS_BLANK(*ptr)) return(NULL);
1585:
1586: while (IS_BLANK(*ptr)) ptr++;
1587: if ((*ptr != '\'') || (*ptr != '"')) return(NULL);
1588: }
1589: SKIP_BLANKS;
1590: URI = htmlParseSystemLiteral(ctxt);
1591: if (URI == NULL) {
1592: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1593: ctxt->sax->error(ctxt->userData,
1594: "htmlParseExternalID: PUBLIC, no URI\n");
1595: ctxt->wellFormed = 0;
1596: }
1597: }
1598: return(URI);
1599: }
1600:
1601: /**
1602: * htmlParseComment:
1603: * @ctxt: an HTML parser context
1604: * @create: should we create a node, or just skip the content
1605: *
1606: * Parse an XML (SGML) comment <!-- .... -->
1607: *
1608: * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->'
1609: */
1610: void
1611: htmlParseComment(htmlParserCtxtPtr ctxt, int create) {
1612: const CHAR *q, *start;
1613: const CHAR *r;
1614: CHAR *val;
1615:
1616: /*
1617: * Check that there is a comment right here.
1618: */
1619: if ((CUR != '<') || (NXT(1) != '!') ||
1620: (NXT(2) != '-') || (NXT(3) != '-')) return;
1621:
1622: SKIP(4);
1623: start = q = CUR_PTR;
1624: NEXT;
1625: r = CUR_PTR;
1626: NEXT;
1627: while (IS_CHAR(CUR) &&
1628: ((CUR == ':') || (CUR != '>') ||
1629: (*r != '-') || (*q != '-'))) {
1630: if ((*r == '-') && (*q == '-')) {
1631: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1632: ctxt->sax->error(ctxt->userData,
1633: "Comment must not contain '--' (double-hyphen)`\n");
1634: ctxt->wellFormed = 0;
1635: }
1636: NEXT;r++;q++;
1637: }
1638: if (!IS_CHAR(CUR)) {
1639: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1640: ctxt->sax->error(ctxt->userData, "Comment not terminated \n<!--%.50s\n", start);
1641: ctxt->wellFormed = 0;
1642: } else {
1643: NEXT;
1644: if (create) {
1645: val = xmlStrndup(start, q - start);
1646: if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL))
1647: ctxt->sax->comment(ctxt->userData, val);
1648: free(val);
1649: }
1650: }
1651: }
1652:
1653: /**
1654: * htmlParseCharRef:
1655: * @ctxt: an HTML parser context
1656: *
1657: * parse Reference declarations
1658: *
1659: * [66] CharRef ::= '&#' [0-9]+ ';' |
1660: * '&#x' [0-9a-fA-F]+ ';'
1661: *
1662: * Returns the value parsed (as an int)
1663: */
1664: int
1665: htmlParseCharRef(htmlParserCtxtPtr ctxt) {
1666: int val = 0;
1667:
1668: if ((CUR == '&') && (NXT(1) == '#') &&
1669: (NXT(2) == 'x')) {
1670: SKIP(3);
1671: while (CUR != ';') {
1672: if ((CUR >= '0') && (CUR <= '9'))
1673: val = val * 16 + (CUR - '0');
1674: else if ((CUR >= 'a') && (CUR <= 'f'))
1675: val = val * 16 + (CUR - 'a') + 10;
1676: else if ((CUR >= 'A') && (CUR <= 'F'))
1677: val = val * 16 + (CUR - 'A') + 10;
1678: else {
1679: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1680: ctxt->sax->error(ctxt->userData,
1681: "htmlParseCharRef: invalid hexadecimal value\n");
1682: ctxt->wellFormed = 0;
1683: val = 0;
1684: break;
1685: }
1686: NEXT;
1687: }
1688: if (CUR == ';')
1689: NEXT;
1690: } else if ((CUR == '&') && (NXT(1) == '#')) {
1691: SKIP(2);
1692: while (CUR != ';') {
1693: if ((CUR >= '0') && (CUR <= '9'))
1694: val = val * 10 + (CUR - '0');
1695: else {
1696: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1697: ctxt->sax->error(ctxt->userData,
1698: "htmlParseCharRef: invalid decimal value\n");
1699: ctxt->wellFormed = 0;
1700: val = 0;
1701: break;
1702: }
1703: NEXT;
1704: }
1705: if (CUR == ';')
1706: NEXT;
1707: } else {
1708: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1709: ctxt->sax->error(ctxt->userData, "htmlParseCharRef: invalid value\n");
1710: ctxt->wellFormed = 0;
1711: }
1712: /*
1713: * Check the value IS_CHAR ...
1714: */
1715: if (IS_CHAR(val)) {
1716: return(val);
1717: } else {
1718: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1719: ctxt->sax->error(ctxt->userData, "htmlParseCharRef: invalid CHAR value %d\n",
1720: val);
1721: ctxt->wellFormed = 0;
1722: }
1723: return(0);
1724: }
1725:
1726:
1727: /**
1728: * htmlParseDocTypeDecl :
1729: * @ctxt: an HTML parser context
1730: *
1731: * parse a DOCTYPE declaration
1732: *
1733: * [28] doctypedecl ::= '<!DOCTYPE' S Name (S ExternalID)? S?
1734: * ('[' (markupdecl | PEReference | S)* ']' S?)? '>'
1735: */
1736:
1737: void
1738: htmlParseDocTypeDecl(htmlParserCtxtPtr ctxt) {
1739: CHAR *name;
1740: CHAR *ExternalID = NULL;
1741: CHAR *URI = NULL;
1742:
1743: /*
1744: * We know that '<!DOCTYPE' has been detected.
1745: */
1746: SKIP(9);
1747:
1748: SKIP_BLANKS;
1749:
1750: /*
1751: * Parse the DOCTYPE name.
1752: */
1753: name = htmlParseName(ctxt);
1754: if (name == NULL) {
1755: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1756: ctxt->sax->error(ctxt->userData, "htmlParseDocTypeDecl : no DOCTYPE name !\n");
1757: ctxt->wellFormed = 0;
1758: }
1759: /*
1760: * Check that upper(name) == "HTML" !!!!!!!!!!!!!
1761: */
1762:
1763: SKIP_BLANKS;
1764:
1765: /*
1766: * Check for SystemID and ExternalID
1767: */
1768: URI = htmlParseExternalID(ctxt, &ExternalID, 1);
1769: SKIP_BLANKS;
1770:
1771: /*
1772: * We should be at the end of the DOCTYPE declaration.
1773: */
1774: if (CUR != '>') {
1775: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1776: ctxt->sax->error(ctxt->userData, "DOCTYPE unproperly terminated\n");
1777: ctxt->wellFormed = 0;
1778: /* We shouldn't try to resynchronize ... */
1779: } else {
1780: }
1781: NEXT;
1782:
1783: /*
1784: * Create the document accordingly to the DOCTYPE
1785: */
1786: ctxt->myDoc = htmlNewDoc(URI, ExternalID);
1787:
1788: /*
1789: * Cleanup, since we don't use all those identifiers
1790: * TODO : the DOCTYPE if available should be stored !
1791: */
1792: if (URI != NULL) free(URI);
1793: if (ExternalID != NULL) free(ExternalID);
1794: if (name != NULL) free(name);
1795: }
1796:
1797: /**
1798: * htmlParseAttribute:
1799: * @ctxt: an HTML parser context
1800: * @value: a CHAR ** used to store the value of the attribute
1801: *
1802: * parse an attribute
1803: *
1804: * [41] Attribute ::= Name Eq AttValue
1805: *
1806: * [25] Eq ::= S? '=' S?
1807: *
1808: * With namespace:
1809: *
1810: * [NS 11] Attribute ::= QName Eq AttValue
1811: *
1812: * Also the case QName == xmlns:??? is handled independently as a namespace
1813: * definition.
1814: *
1815: * Returns the attribute name, and the value in *value.
1816: */
1817:
1818: CHAR *
1819: htmlParseAttribute(htmlParserCtxtPtr ctxt, CHAR **value) {
1820: CHAR *name, *val;
1821:
1822: *value = NULL;
1823: name = htmlParseName(ctxt);
1824: if (name == NULL) {
1825: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1826: ctxt->sax->error(ctxt->userData, "error parsing attribute name\n");
1827: ctxt->wellFormed = 0;
1828: return(NULL);
1829: }
1830:
1831: /*
1832: * read the value
1833: */
1834: SKIP_BLANKS;
1835: if (CUR == '=') {
1836: NEXT;
1837: SKIP_BLANKS;
1838: val = htmlParseAttValue(ctxt);
1839: } else {
1840: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1841: ctxt->sax->error(ctxt->userData,
1842: "Specification mandate value for attribute %s\n", name);
1843: ctxt->wellFormed = 0;
1844: return(NULL);
1845: }
1846:
1847: *value = val;
1848: return(name);
1849: }
1850:
1851: /**
1852: * htmlParseStartTag:
1853: * @ctxt: an HTML parser context
1854: *
1855: * parse a start of tag either for rule element or
1856: * EmptyElement. In both case we don't parse the tag closing chars.
1857: *
1858: * [40] STag ::= '<' Name (S Attribute)* S? '>'
1859: *
1860: * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>'
1861: *
1862: * With namespace:
1863: *
1864: * [NS 8] STag ::= '<' QName (S Attribute)* S? '>'
1865: *
1866: * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>'
1867: *
1868: * Returns the element name parsed
1869: */
1870:
1871: CHAR *
1872: htmlParseStartTag(htmlParserCtxtPtr ctxt) {
1873: CHAR *name;
1874: CHAR *attname;
1875: CHAR *attvalue;
1876: const CHAR **atts = NULL;
1877: int nbatts = 0;
1878: int maxatts = 0;
1879: int i;
1880:
1881: if (CUR != '<') return(NULL);
1882: NEXT;
1883:
1884: name = htmlParseHTMLName(ctxt);
1885: if (name == NULL) {
1886: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1887: ctxt->sax->error(ctxt->userData,
1888: "htmlParseStartTag: invalid element name\n");
1889: ctxt->wellFormed = 0;
1890: return(NULL);
1891: }
1892:
1893: /*
1894: * Check for auto-closure of HTML elements.
1895: */
1896: htmlAutoClose(ctxt, name);
1897:
1898: /*
1899: * Now parse the attributes, it ends up with the ending
1900: *
1901: * (S Attribute)* S?
1902: */
1903: SKIP_BLANKS;
1904: while ((IS_CHAR(CUR)) &&
1905: (CUR != '>') &&
1906: ((CUR != '/') || (NXT(1) != '>'))) {
1907: const CHAR *q = CUR_PTR;
1908:
1909: attname = htmlParseAttribute(ctxt, &attvalue);
1910: if ((attname != NULL) && (attvalue != NULL)) {
1911: /*
1912: * Well formedness requires at most one declaration of an attribute
1913: */
1914: for (i = 0; i < nbatts;i += 2) {
1915: if (!xmlStrcmp(atts[i], attname)) {
1916: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1917: ctxt->sax->error(ctxt->userData, "Attribute %s redefined\n",
1918: name);
1919: ctxt->wellFormed = 0;
1920: free(attname);
1921: free(attvalue);
1922: break;
1923: }
1924: }
1925:
1926: /*
1927: * Add the pair to atts
1928: */
1929: if (atts == NULL) {
1930: maxatts = 10;
1931: atts = (const CHAR **) malloc(maxatts * sizeof(CHAR *));
1932: if (atts == NULL) {
1933: fprintf(stderr, "malloc of %ld byte failed\n",
1934: maxatts * (long)sizeof(CHAR *));
1935: return(NULL);
1936: }
1937: } else if (nbatts + 2 < maxatts) {
1938: maxatts *= 2;
1939: atts = (const CHAR **) realloc(atts, maxatts * sizeof(CHAR *));
1940: if (atts == NULL) {
1941: fprintf(stderr, "realloc of %ld byte failed\n",
1942: maxatts * (long)sizeof(CHAR *));
1943: return(NULL);
1944: }
1945: }
1946: atts[nbatts++] = attname;
1947: atts[nbatts++] = attvalue;
1948: atts[nbatts] = NULL;
1949: atts[nbatts + 1] = NULL;
1950: }
1951:
1952: SKIP_BLANKS;
1953: if (q == CUR_PTR) {
1954: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1955: ctxt->sax->error(ctxt->userData,
1956: "htmlParseStartTag: problem parsing attributes\n");
1957: ctxt->wellFormed = 0;
1958: break;
1959: }
1960: }
1961:
1962: /*
1963: * SAX: Start of Element !
1964: */
1965: if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL))
1966: ctxt->sax->startElement(ctxt->userData, name, atts);
1967:
1968: if (atts != NULL) {
1969: for (i = 0;i < nbatts;i++) free((CHAR *) atts[i]);
1970: free(atts);
1971: }
1972: return(name);
1973: }
1974:
1975: /**
1976: * htmlParseEndTag:
1977: * @ctxt: an HTML parser context
1978: * @tagname: the tag name as parsed in the opening tag.
1979: *
1980: * parse an end of tag
1981: *
1982: * [42] ETag ::= '</' Name S? '>'
1983: *
1984: * With namespace
1985: *
1986: * [NS 9] ETag ::= '</' QName S? '>'
1987: */
1988:
1989: void
1990: htmlParseEndTag(htmlParserCtxtPtr ctxt, const CHAR *tagname) {
1991: CHAR *name;
1992: int i;
1993:
1994: if ((CUR != '<') || (NXT(1) != '/')) {
1995: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
1996: ctxt->sax->error(ctxt->userData, "htmlParseEndTag: '</' not found\n");
1997: ctxt->wellFormed = 0;
1998: return;
1999: }
2000: SKIP(2);
2001:
2002: name = htmlParseHTMLName(ctxt);
2003:
2004: /*
2005: * We should definitely be at the ending "S? '>'" part
2006: */
2007: SKIP_BLANKS;
2008: if ((!IS_CHAR(CUR)) || (CUR != '>')) {
2009: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2010: ctxt->sax->error(ctxt->userData, "End tag : expected '>'\n");
2011: ctxt->wellFormed = 0;
2012: } else
2013: NEXT;
2014:
2015: /*
2016: * Check that we are not closing an already closed tag,
2017: * <p><b>...</p></b> is a really common error !
2018: */
2019: for (i = ctxt->nodeNr - 1;i >= 0;i--) {
2020: if ((ctxt->nodeTab[i] != NULL) &&
2021: (!xmlStrcmp(tagname, ctxt->nodeTab[i]->name)))
2022: break;
2023: }
2024: if (i < 0) {
2025: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2026: ctxt->sax->error(ctxt->userData,
2027: "htmlParseEndTag: unexpected close for tag %s\n",
2028: tagname);
2029: ctxt->wellFormed = 0;
2030: return;
2031: }
2032:
2033: /*
2034: * Check for auto-closure of HTML elements.
2035: */
2036: htmlAutoCloseOnClose(ctxt, name);
2037:
2038: /*
2039: * Well formedness constraints, opening and closing must match.
2040: * With the exception that the autoclose may have popped stuff out
2041: * of the stack.
2042: */
2043: if (xmlStrcmp(name, tagname)) {
2044: if ((ctxt->node != NULL) &&
2045: (xmlStrcmp(ctxt->node->name, name))) {
2046: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2047: ctxt->sax->error(ctxt->userData,
2048: "Opening and ending tag mismatch: %s and %s\n",
2049: name, ctxt->node->name);
2050: ctxt->wellFormed = 0;
2051: }
2052: }
2053:
2054: /*
2055: * SAX: End of Tag
2056: */
2057: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
2058: ctxt->sax->endElement(ctxt->userData, name);
2059:
2060: if (name != NULL)
2061: free(name);
2062:
2063: return;
2064: }
2065:
2066:
2067: /**
2068: * htmlParseReference:
2069: * @ctxt: an HTML parser context
2070: *
2071: * parse and handle entity references in content,
2072: * this will end-up in a call to character() since this is either a
2073: * CharRef, or a predefined entity.
2074: */
2075: void
2076: htmlParseReference(htmlParserCtxtPtr ctxt) {
2077: htmlEntityDescPtr ent;
2078: CHAR out[2];
2079: CHAR *name;
2080: int val;
2081: if (CUR != '&') return;
2082:
2083: if (NXT(1) == '#') {
2084: val = htmlParseCharRef(ctxt);
2085: /* TODO: invalid for UTF-8 variable encoding !!! */
2086: out[0] = val;
2087: out[1] = 0;
2088: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
2089: ctxt->sax->characters(ctxt->userData, out, 1);
2090: } else {
2091: ent = htmlParseEntityRef(ctxt, &name);
2092: if (name == NULL) return; /* Shall we output & anyway ? */
2093: if ((ent == NULL) || (ent->value <= 0) || (ent->value >= 255)) {
2094: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) {
2095: ctxt->sax->characters(ctxt->userData, "&", 1);
2096: ctxt->sax->characters(ctxt->userData, name, xmlStrlen(name));
2097: ctxt->sax->characters(ctxt->userData, ";", 1);
2098: }
2099: } else {
2100: /* TODO: invalid for UTF-8 variable encoding !!! */
2101: out[0] = ent->value;
2102: out[1] = 0;
2103: if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL))
2104: ctxt->sax->characters(ctxt->userData, out, 1);
2105: }
2106: free(name);
2107: }
2108: }
2109:
2110: /**
2111: * htmlParseContent:
2112: * @ctxt: an HTML parser context
2113: * @name: the node name
2114: *
2115: * Parse a content: comment, sub-element, reference or text.
2116: *
2117: */
2118:
2119: void
2120: htmlParseContent(htmlParserCtxtPtr ctxt, const CHAR *name) {
2121: htmlNodePtr currentNode;
2122:
2123: currentNode = ctxt->node;
2124: while ((CUR != '<') || (NXT(1) != '/')) {
2125: const CHAR *test = CUR_PTR;
2126:
2127: /*
2128: * Has this node been popped out during parsing of
2129: * the next element
2130: */
2131: if (currentNode != ctxt->node) return;
2132:
2133: /*
2134: * First case : a comment
2135: */
2136: if ((CUR == '<') && (NXT(1) == '!') &&
2137: (NXT(2) == '-') && (NXT(3) == '-')) {
2138: htmlParseComment(ctxt, 1);
2139: }
2140:
2141: /*
2142: * Second case : a sub-element.
2143: */
2144: else if (CUR == '<') {
2145: htmlParseElement(ctxt);
2146: }
2147:
2148: /*
2149: * Third case : a reference. If if has not been resolved,
2150: * parsing returns it's Name, create the node
2151: */
2152: else if (CUR == '&') {
2153: htmlParseReference(ctxt);
2154: }
2155:
2156: /*
2157: * Last case, text. Note that References are handled directly.
2158: */
2159: else {
2160: htmlParseCharData(ctxt, 0);
2161: }
2162:
2163: if (test == CUR_PTR) {
2164: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2165: ctxt->sax->error(ctxt->userData,
2166: "detected an error in element content\n");
2167: ctxt->wellFormed = 0;
2168: break;
2169: }
2170: }
2171:
2172: /*
2173: * parse the end of tag: '</' should be here.
2174: */
2175: htmlParseEndTag(ctxt, name);
2176: }
2177:
2178: /**
2179: * htmlParseElement:
2180: * @ctxt: an HTML parser context
2181: *
2182: * parse an HTML element, this is highly recursive
2183: *
2184: * [39] element ::= EmptyElemTag | STag content ETag
2185: *
2186: * [41] Attribute ::= Name Eq AttValue
2187: */
2188:
2189: void
2190: htmlParseElement(htmlParserCtxtPtr ctxt) {
2191: const CHAR *openTag = CUR_PTR;
2192: CHAR *name;
2193: htmlParserNodeInfo node_info;
2194: htmlNodePtr currentNode;
2195: htmlElemDescPtr info;
2196:
2197: /* Capture start position */
2198: node_info.begin_pos = CUR_PTR - ctxt->input->base;
2199: node_info.begin_line = ctxt->input->line;
2200:
2201: name = htmlParseStartTag(ctxt);
2202: if (name == NULL) {
2203: return;
2204: }
2205:
2206: /*
2207: * Lookup the info for that element.
2208: */
2209: info = htmlTagLookup(name);
2210: if (info == NULL) {
2211: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2212: ctxt->sax->error(ctxt->userData, "Tag %s invalid\n",
2213: name);
2214: ctxt->wellFormed = 0;
2215: } else if (info->depr) {
2216: /***************************
2217: if ((ctxt->sax != NULL) && (ctxt->sax->warning != NULL))
2218: ctxt->sax->warning(ctxt->userData, "Tag %s is deprecated\n",
2219: name);
2220: ***************************/
2221: }
2222:
2223: /*
2224: * Check for an Empty Element labelled the XML/SGML way
2225: */
2226: if ((CUR == '/') && (NXT(1) == '>')) {
2227: SKIP(2);
2228: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
2229: ctxt->sax->endElement(ctxt->userData, name);
2230: free(name);
2231: return;
2232: }
2233:
2234: if (CUR == '>') NEXT;
2235: else {
2236: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2237: ctxt->sax->error(ctxt->userData, "Couldn't find end of Start Tag\n%.30s\n",
2238: openTag);
2239: ctxt->wellFormed = 0;
2240:
2241: /*
2242: * end of parsing of this node.
2243: */
2244: nodePop(ctxt);
2245: free(name);
2246: return;
2247: }
2248:
2249: /*
2250: * Check for an Empty Element from DTD definition
2251: */
2252: if ((info != NULL) && (info->empty)) {
2253: if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL))
2254: ctxt->sax->endElement(ctxt->userData, name);
2255: free(name);
2256: return;
2257: }
2258:
2259: /*
2260: * Parse the content of the element:
2261: */
2262: currentNode = ctxt->node;
2263: htmlParseContent(ctxt, name);
2264:
2265: /*
2266: * check whether the element get popped due to auto closure
2267: * on start tag
2268: */
2269: if (currentNode != ctxt->node) {
2270: free(name);
2271: return;
2272: }
2273:
2274: if (!IS_CHAR(CUR)) {
2275: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2276: ctxt->sax->error(ctxt->userData,
2277: "Premature end of data in tag %.30s\n", openTag);
2278: ctxt->wellFormed = 0;
2279:
2280: /*
2281: * end of parsing of this node.
2282: */
2283: nodePop(ctxt);
2284: free(name);
2285: return;
2286: }
2287:
2288: free(name);
2289: }
2290:
2291: /**
2292: * htmlParseDocument :
2293: * @ctxt: an HTML parser context
2294: *
2295: * parse an HTML document (and build a tree if using the standard SAX
2296: * interface).
2297: *
2298: * Returns 0, -1 in case of error. the parser context is augmented
2299: * as a result of the parsing.
2300: */
2301:
2302: int
2303: htmlParseDocument(htmlParserCtxtPtr ctxt) {
2304: htmlDefaultSAXHandlerInit();
2305: ctxt->html = 1;
2306:
2307: /*
2308: * SAX: beginning of the document processing TODO: update for HTML.
2309: */
2310: if ((ctxt->sax) && (ctxt->sax->setDocumentLocator))
2311: ctxt->sax->setDocumentLocator(ctxt->userData, &xmlDefaultSAXLocator);
2312:
2313: /*
2314: * We should check for encoding here and plug-in some
2315: * conversion code TODO !!!!
2316: */
2317:
2318: /*
2319: * Wipe out everything which is before the first '<'
2320: */
2321: if (IS_BLANK(CUR)) {
2322: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2323: ctxt->sax->error(ctxt->userData,
2324: "Extra spaces at the beginning of the document are not allowed\n");
2325: ctxt->wellFormed = 0;
2326: SKIP_BLANKS;
2327: }
2328:
2329: if (CUR == 0) {
2330: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
2331: ctxt->sax->error(ctxt->userData, "Document is empty\n");
2332: ctxt->wellFormed = 0;
2333: }
2334:
2335:
2336: /*
2337: * Then possibly doc type declaration(s) and more Misc
2338: * (doctypedecl Misc*)?
2339: */
2340: if ((CUR == '<') && (NXT(1) == '!') &&
2341: (UPP(2) == 'D') && (UPP(3) == 'O') &&
2342: (UPP(4) == 'C') && (UPP(5) == 'T') &&
2343: (UPP(6) == 'Y') && (UPP(7) == 'P') &&
2344: (UPP(8) == 'E')) {
2345: htmlParseDocTypeDecl(ctxt);
2346: }
2347: SKIP_BLANKS;
2348:
2349: /*
2350: * Create the document if not done already.
2351: */
2352: if (ctxt->myDoc == NULL) {
2353: ctxt->myDoc = htmlNewDoc(NULL, NULL);
2354: }
2355:
2356: /*
2357: * Time to start parsing the tree itself
2358: */
2359: htmlParseElement(ctxt);
2360:
2361: /*
2362: * SAX: end of the document processing.
2363: */
2364: if ((ctxt->sax) && (ctxt->sax->endDocument != NULL))
2365: ctxt->sax->endDocument(ctxt->userData);
2366: if (! ctxt->wellFormed) return(-1);
2367: return(0);
2368: }
2369:
2370:
2371: /********************************************************************************
2372: * *
2373: * Parser contexts handling *
2374: * *
2375: ********************************************************************************/
2376:
2377: /**
2378: * xmlInitParserCtxt:
2379: * @ctxt: an HTML parser context
2380: *
2381: * Initialize a parser context
2382: */
2383:
2384: void
2385: htmlInitParserCtxt(htmlParserCtxtPtr ctxt)
2386: {
2387: htmlSAXHandler *sax;
2388:
2389: sax = (htmlSAXHandler *) malloc(sizeof(htmlSAXHandler));
2390: if (sax == NULL) {
2391: fprintf(stderr, "htmlInitParserCtxt: out of memory\n");
2392: }
2393:
2394: /* Allocate the Input stack */
2395: ctxt->inputTab = (htmlParserInputPtr *) malloc(5 * sizeof(htmlParserInputPtr));
2396: ctxt->inputNr = 0;
2397: ctxt->inputMax = 5;
2398: ctxt->input = NULL;
2399: ctxt->version = NULL;
2400: ctxt->encoding = NULL;
2401: ctxt->standalone = -1;
2402:
2403: /* Allocate the Node stack */
2404: ctxt->nodeTab = (htmlNodePtr *) malloc(10 * sizeof(htmlNodePtr));
2405: ctxt->nodeNr = 0;
2406: ctxt->nodeMax = 10;
2407: ctxt->node = NULL;
2408:
2409: if (sax == NULL) ctxt->sax = &htmlDefaultSAXHandler;
2410: else {
2411: ctxt->sax = sax;
2412: memcpy(sax, &htmlDefaultSAXHandler, sizeof(htmlSAXHandler));
2413: }
2414: ctxt->userData = ctxt;
2415: ctxt->myDoc = NULL;
2416: ctxt->wellFormed = 1;
2417: ctxt->replaceEntities = 0;
2418: ctxt->html = 1;
2419: ctxt->record_info = 0;
2420: xmlInitNodeInfoSeq(&ctxt->node_seq);
2421: }
2422:
2423: /**
2424: * htmlFreeParserCtxt:
2425: * @ctxt: an HTML parser context
2426: *
2427: * Free all the memory used by a parser context. However the parsed
2428: * document in ctxt->myDoc is not freed.
2429: */
2430:
2431: void
2432: htmlFreeParserCtxt(htmlParserCtxtPtr ctxt)
2433: {
2434: htmlParserInputPtr input;
2435:
2436: if (ctxt == NULL) return;
2437:
2438: while ((input = inputPop(ctxt)) != NULL) {
2439: xmlFreeInputStream(input);
2440: }
2441:
2442: if (ctxt->nodeTab != NULL) free(ctxt->nodeTab);
2443: if (ctxt->inputTab != NULL) free(ctxt->inputTab);
2444: if (ctxt->version != NULL) free((char *) ctxt->version);
2445: if ((ctxt->sax != NULL) && (ctxt->sax != &htmlDefaultSAXHandler))
2446: free(ctxt->sax);
2447: free(ctxt);
2448: }
2449:
2450: /**
2451: * htmlCreateDocParserCtxt :
2452: * @cur: a pointer to an array of CHAR
2453: * @encoding: a free form C string describing the HTML document encoding, or NULL
2454: *
2455: * Create a parser context for an HTML document.
2456: *
2457: * Returns the new parser context or NULL
2458: */
2459: htmlParserCtxtPtr
2460: htmlCreateDocParserCtxt(CHAR *cur, const char *encoding) {
2461: htmlParserCtxtPtr ctxt;
2462: htmlParserInputPtr input;
2463: /* htmlCharEncoding enc; */
2464:
2465: ctxt = (htmlParserCtxtPtr) malloc(sizeof(htmlParserCtxt));
2466: if (ctxt == NULL) {
2467: perror("malloc");
2468: return(NULL);
2469: }
2470: htmlInitParserCtxt(ctxt);
2471: input = (htmlParserInputPtr) malloc(sizeof(htmlParserInput));
2472: if (input == NULL) {
2473: perror("malloc");
2474: free(ctxt);
2475: return(NULL);
2476: }
2477:
2478: /*
2479: * plug some encoding conversion routines here. !!!
2480: if (encoding != NULL) {
2481: enc = htmlDetectCharEncoding(cur);
2482: htmlSwitchEncoding(ctxt, enc);
2483: }
2484: */
2485:
2486: input->filename = NULL;
2487: input->line = 1;
2488: input->col = 1;
2489: input->base = cur;
2490: input->cur = cur;
2491: input->free = NULL;
2492:
2493: inputPush(ctxt, input);
2494: return(ctxt);
2495: }
2496:
2497: /********************************************************************************
2498: * *
2499: * User entry points *
2500: * *
2501: ********************************************************************************/
2502:
2503: /**
2504: * htmlSAXParseDoc :
2505: * @cur: a pointer to an array of CHAR
2506: * @encoding: a free form C string describing the HTML document encoding, or NULL
2507: * @sax: the SAX handler block
2508: * @userData: if using SAX, this pointer will be provided on callbacks.
2509: *
2510: * parse an HTML in-memory document and build a tree.
2511: * It use the given SAX function block to handle the parsing callback.
2512: * If sax is NULL, fallback to the default DOM tree building routines.
2513: *
2514: * Returns the resulting document tree
2515: */
2516:
2517: htmlDocPtr
2518: htmlSAXParseDoc(CHAR *cur, const char *encoding, htmlSAXHandlerPtr sax, void *userData) {
2519: htmlDocPtr ret;
2520: htmlParserCtxtPtr ctxt;
2521:
2522: if (cur == NULL) return(NULL);
2523:
2524:
2525: ctxt = htmlCreateDocParserCtxt(cur, encoding);
2526: if (ctxt == NULL) return(NULL);
2527: if (sax != NULL) {
2528: ctxt->sax = sax;
2529: ctxt->userData = userData;
2530: }
2531:
2532: htmlParseDocument(ctxt);
2533: ret = ctxt->myDoc;
2534: if (sax != NULL) {
2535: ctxt->sax = NULL;
2536: ctxt->userData = NULL;
2537: }
2538: htmlFreeParserCtxt(ctxt);
2539:
2540: return(ret);
2541: }
2542:
2543: /**
2544: * htmlParseDoc :
2545: * @cur: a pointer to an array of CHAR
2546: * @encoding: a free form C string describing the HTML document encoding, or NULL
2547: *
2548: * parse an HTML in-memory document and build a tree.
2549: *
2550: * Returns the resulting document tree
2551: */
2552:
2553: htmlDocPtr
2554: htmlParseDoc(CHAR *cur, const char *encoding) {
2555: return(htmlSAXParseDoc(cur, encoding, NULL, NULL));
2556: }
2557:
2558:
2559: /**
2560: * htmlCreateFileParserCtxt :
2561: * @filename: the filename
2562: * @encoding: a free form C string describing the HTML document encoding, or NULL
2563: *
2564: * Create a parser context for a file content.
2565: * Automatic support for ZLIB/Compress compressed document is provided
2566: * by default if found at compile-time.
2567: *
2568: * Returns the new parser context or NULL
2569: */
2570: htmlParserCtxtPtr
2571: htmlCreateFileParserCtxt(const char *filename, const char *encoding)
2572: {
2573: htmlParserCtxtPtr ctxt;
2574: #ifdef HAVE_ZLIB_H
2575: gzFile input;
2576: #else
2577: int input;
2578: #endif
2579: int res;
2580: int len;
2581: int cnt;
2582: struct stat buf;
2583: char *buffer, *nbuf;
2584: htmlParserInputPtr inputStream;
2585: /* htmlCharEncoding enc; */
2586:
2587: #define MINLEN 40000
2588:
2589: if (strcmp(filename,"-") == 0) {
2590: #ifdef HAVE_ZLIB_H
2591: input = gzdopen (fileno(stdin), "r");
2592: if (input == NULL) {
2593: fprintf (stderr, "Cannot read from stdin\n");
2594: perror ("gzdopen failed");
2595: return(NULL);
2596: }
2597: #else
2598: #ifdef WIN32
2599: input = -1;
2600: #else
2601: input = fileno(stdin);
2602: #endif
2603: if (input < 0) {
2604: fprintf (stderr, "Cannot read from stdin\n");
2605: perror ("open failed");
2606: return(NULL);
2607: }
2608: #endif
2609: len = MINLEN;
2610: } else {
2611: #ifdef HAVE_ZLIB_H
2612: input = gzopen (filename, "r");
2613: if (input == NULL) {
2614: fprintf (stderr, "Cannot read file %s :\n", filename);
2615: perror ("gzopen failed");
2616: return(NULL);
2617: }
2618: #else
2619: #ifdef WIN32
2620: input = _open (filename, O_RDONLY | _O_BINARY);
2621: #else
2622: input = open (filename, O_RDONLY);
2623: #endif
2624: if (input < 0) {
2625: fprintf (stderr, "Cannot read file %s :\n", filename);
2626: perror ("open failed");
2627: return(NULL);
2628: }
2629: #endif
2630: res = stat(filename, &buf);
2631: if (res < 0)
2632: return(NULL);
1.2 daniel 2633: len = buf.st_size;
1.1 daniel 2634: if (len < MINLEN)
2635: len = MINLEN;
2636: }
1.2 daniel 2637: buffer = (char *)malloc((len+1)*sizeof(char));
1.1 daniel 2638: if (buffer == NULL) {
2639: fprintf (stderr, "Cannot malloc\n");
2640: perror ("malloc failed");
2641: return(NULL);
2642: }
2643:
2644: cnt = 0;
2645: while(1) {
2646: if (cnt == len) {
2647: len *= 2;
1.2 daniel 2648: nbuf = (char *)realloc(buffer,(len+1)*sizeof(char));
1.1 daniel 2649: if (nbuf == NULL) {
2650: fprintf(stderr,"Cannot realloc\n");
2651: free(buffer);
2652: perror ("realloc failed");
2653: return(NULL);
2654: }
2655: buffer = nbuf;
2656: }
2657: #ifdef HAVE_ZLIB_H
2658: res = gzread(input, &buffer[cnt], len-cnt);
2659: #else
2660: res = read(input, &buffer[cnt], len-cnt);
2661: #endif
2662: if (res < 0) {
2663: fprintf (stderr, "Cannot read file %s :\n", filename);
2664: #ifdef HAVE_ZLIB_H
2665: perror ("gzread failed");
2666: #else
2667: perror ("read failed");
2668: #endif
2669: return(NULL);
2670: }
2671: if (res == 0)
2672: break;
2673: cnt +=res;
2674: }
2675: buffer[cnt] = '\0';
2676:
2677: #ifdef HAVE_ZLIB_H
2678: gzclose(input);
2679: #else
2680: close(input);
2681: #endif
2682:
2683: ctxt = (htmlParserCtxtPtr) malloc(sizeof(htmlParserCtxt));
2684: if (ctxt == NULL) {
2685: perror("malloc");
2686: return(NULL);
2687: }
2688: htmlInitParserCtxt(ctxt);
2689: inputStream = (htmlParserInputPtr) malloc(sizeof(htmlParserInput));
2690: if (inputStream == NULL) {
2691: perror("malloc");
2692: free(ctxt);
2693: return(NULL);
2694: }
2695:
2696: inputStream->filename = strdup(filename);
2697: inputStream->line = 1;
2698: inputStream->col = 1;
2699:
2700: /*
2701: * plug some encoding conversion routines here. !!!
2702: if (encoding != NULL) {
2703: enc = htmlDetectCharEncoding(buffer);
2704: htmlSwitchEncoding(ctxt, enc);
2705: }
2706: */
2707:
2708: inputStream->base = buffer;
2709: inputStream->cur = buffer;
2710: inputStream->free = (xmlParserInputDeallocate) free;
2711:
2712: inputPush(ctxt, inputStream);
2713: return(ctxt);
2714: }
2715:
2716: /**
2717: * htmlSAXParseFile :
2718: * @filename: the filename
2719: * @encoding: a free form C string describing the HTML document encoding, or NULL
2720: * @sax: the SAX handler block
2721: * @userData: if using SAX, this pointer will be provided on callbacks.
2722: *
2723: * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
2724: * compressed document is provided by default if found at compile-time.
2725: * It use the given SAX function block to handle the parsing callback.
2726: * If sax is NULL, fallback to the default DOM tree building routines.
2727: *
2728: * Returns the resulting document tree
2729: */
2730:
2731: htmlDocPtr
2732: htmlSAXParseFile(const char *filename, const char *encoding, htmlSAXHandlerPtr sax,
2733: void *userData) {
2734: htmlDocPtr ret;
2735: htmlParserCtxtPtr ctxt;
2736:
2737: ctxt = htmlCreateFileParserCtxt(filename, encoding);
2738: if (ctxt == NULL) return(NULL);
2739: if (sax != NULL) {
2740: ctxt->sax = sax;
2741: ctxt->userData = userData;
2742: }
2743:
2744: htmlParseDocument(ctxt);
2745:
2746: ret = ctxt->myDoc;
2747: if (sax != NULL) {
2748: ctxt->sax = NULL;
2749: ctxt->userData = NULL;
2750: }
2751: htmlFreeParserCtxt(ctxt);
2752:
2753: return(ret);
2754: }
2755:
2756: /**
2757: * htmlParseFile :
2758: * @filename: the filename
2759: * @encoding: a free form C string describing the HTML document encoding, or NULL
2760: *
2761: * parse an HTML file and build a tree. Automatic support for ZLIB/Compress
2762: * compressed document is provided by default if found at compile-time.
2763: *
2764: * Returns the resulting document tree
2765: */
2766:
2767: htmlDocPtr
2768: htmlParseFile(const char *filename, const char *encoding) {
2769: return(htmlSAXParseFile(filename, encoding, NULL, NULL));
2770: }
Webmaster
![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to HTMLparser.c CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [Public] / XML